id	sid	tid	token	lemma	pos
ijassa-348	1	1	advances	advance	NOUN
ijassa-348	1	2	in	in	ADP
ijassa-348	1	3	systems	system	NOUN
ijassa-348	1	4	science	science	NOUN
ijassa-348	1	5	and	and	CCONJ
ijassa-348	1	6	application	application	NOUN
ijassa-348	1	7	(	(	PUNCT
ijassa-348	1	8	2016	2016	NUM
ijassa-348	1	9	)	)	PUNCT
ijassa-348	1	10	vol.16	vol.16	PROPN
ijassa-348	2	1	no.2	no.2	PROPN
ijassa-348	2	2	15	15	NUM
ijassa-348	2	3	-	-	SYM
ijassa-348	2	4	38	38	NUM
ijassa-348	2	5	intrusion	intrusion	NOUN
ijassa-348	2	6	detection	detection	NOUN
ijassa-348	2	7	model	model	NOUN
ijassa-348	2	8	using	use	VERB
ijassa-348	2	9	pca	pca	PROPN
ijassa-348	2	10	and	and	CCONJ
ijassa-348	2	11	ensemble	ensemble	ADJ
ijassa-348	2	12	of	of	ADP
ijassa-348	2	13	classifiers	classifier	NOUN
ijassa-348	2	14	sumaiya	sumaiya	VERB
ijassa-348	2	15	thaseen1	thaseen1	ADJ
ijassa-348	2	16	and	and	CCONJ
ijassa-348	2	17	ch.aswani	ch.aswani	X
ijassa-348	2	18	kumar2	kumar2	X
ijassa-348	3	1	1school	1school	NUM
ijassa-348	3	2	of	of	ADP
ijassa-348	3	3	computing	compute	VERB
ijassa-348	3	4	science	science	NOUN
ijassa-348	3	5	and	and	CCONJ
ijassa-348	3	6	engineering	engineering	NOUN
ijassa-348	3	7	,	,	PUNCT
ijassa-348	3	8	vit	vit	PROPN
ijassa-348	3	9	university	university	NOUN
ijassa-348	3	10	,	,	PUNCT
ijassa-348	3	11	chennai	chennai	PROPN
ijassa-348	3	12	,	,	PUNCT
ijassa-348	3	13	tamil	tamil	PROPN
ijassa-348	3	14	nadu	nadu	NOUN
ijassa-348	3	15	.	.	PUNCT
ijassa-348	4	1	2school	2school	NUM
ijassa-348	4	2	of	of	ADP
ijassa-348	4	3	information	information	NOUN
ijassa-348	4	4	technology	technology	NOUN
ijassa-348	4	5	and	and	CCONJ
ijassa-348	4	6	engineering	engineering	NOUN
ijassa-348	4	7	,	,	PUNCT
ijassa-348	4	8	vit	vit	PROPN
ijassa-348	4	9	university	university	NOUN
ijassa-348	4	10	,	,	PUNCT
ijassa-348	4	11	vellore	vellore	NOUN
ijassa-348	4	12	.	.	PUNCT
ijassa-348	5	1	abstract	abstract	ADV
ijassa-348	5	2	most	most	ADJ
ijassa-348	5	3	of	of	ADP
ijassa-348	5	4	the	the	DET
ijassa-348	5	5	intrusion	intrusion	NOUN
ijassa-348	5	6	detection	detection	NOUN
ijassa-348	5	7	systems	system	NOUN
ijassa-348	5	8	examine	examine	VERB
ijassa-348	5	9	all	all	DET
ijassa-348	5	10	network	network	NOUN
ijassa-348	5	11	features	feature	VERB
ijassa-348	5	12	to	to	PART
ijassa-348	5	13	identify	identify	VERB
ijassa-348	5	14	intrusions	intrusion	NOUN
ijassa-348	5	15	with	with	ADP
ijassa-348	5	16	different	different	ADJ
ijassa-348	5	17	classification	classification	NOUN
ijassa-348	5	18	approaches	approach	NOUN
ijassa-348	5	19	.	.	PUNCT
ijassa-348	6	1	the	the	DET
ijassa-348	6	2	major	major	ADJ
ijassa-348	6	3	challenges	challenge	NOUN
ijassa-348	6	4	for	for	ADP
ijassa-348	6	5	any	any	DET
ijassa-348	6	6	intrusion	intrusion	NOUN
ijassa-348	6	7	detection	detection	NOUN
ijassa-348	6	8	model	model	NOUN
ijassa-348	6	9	is	be	AUX
ijassa-348	6	10	to	to	PART
ijassa-348	6	11	achieve	achieve	VERB
ijassa-348	6	12	maximum	maximum	ADJ
ijassa-348	6	13	accuracy	accuracy	NOUN
ijassa-348	6	14	with	with	ADP
ijassa-348	6	15	minimal	minimal	ADJ
ijassa-348	6	16	false	false	ADJ
ijassa-348	6	17	alarms	alarm	NOUN
ijassa-348	6	18	.	.	PUNCT
ijassa-348	7	1	while	while	SCONJ
ijassa-348	7	2	many	many	ADJ
ijassa-348	7	3	ensemble	ensemble	ADJ
ijassa-348	7	4	techniques	technique	NOUN
ijassa-348	7	5	are	be	AUX
ijassa-348	7	6	present	present	ADJ
ijassa-348	7	7	to	to	PART
ijassa-348	7	8	improve	improve	VERB
ijassa-348	7	9	the	the	DET
ijassa-348	7	10	accuracy	accuracy	NOUN
ijassa-348	7	11	of	of	ADP
ijassa-348	7	12	intrusion	intrusion	NOUN
ijassa-348	7	13	detection	detection	NOUN
ijassa-348	7	14	models	model	NOUN
ijassa-348	7	15	,	,	PUNCT
ijassa-348	7	16	building	build	VERB
ijassa-348	7	17	an	an	DET
ijassa-348	7	18	ensemble	ensemble	ADJ
ijassa-348	7	19	that	that	PRON
ijassa-348	7	20	can	can	AUX
ijassa-348	7	21	be	be	AUX
ijassa-348	7	22	generically	generically	ADV
ijassa-348	7	23	applied	apply	VERB
ijassa-348	7	24	for	for	ADP
ijassa-348	7	25	any	any	DET
ijassa-348	7	26	network	network	NOUN
ijassa-348	7	27	traffic	traffic	NOUN
ijassa-348	7	28	is	be	AUX
ijassa-348	7	29	still	still	ADV
ijassa-348	7	30	a	a	DET
ijassa-348	7	31	difficult	difficult	ADJ
ijassa-348	7	32	task	task	NOUN
ijassa-348	7	33	.	.	PUNCT
ijassa-348	8	1	in	in	ADP
ijassa-348	8	2	this	this	DET
ijassa-348	8	3	paper	paper	NOUN
ijassa-348	8	4	,	,	PUNCT
ijassa-348	8	5	we	we	PRON
ijassa-348	8	6	propose	propose	VERB
ijassa-348	8	7	a	a	DET
ijassa-348	8	8	hybrid	hybrid	ADJ
ijassa-348	8	9	model	model	NOUN
ijassa-348	8	10	for	for	ADP
ijassa-348	8	11	intrusion	intrusion	NOUN
ijassa-348	8	12	detection	detection	NOUN
ijassa-348	8	13	integrating	integrate	VERB
ijassa-348	8	14	base	base	NOUN
ijassa-348	8	15	classifiers	classifier	NOUN
ijassa-348	8	16	such	such	ADJ
ijassa-348	8	17	as	as	ADP
ijassa-348	8	18	svm	svm	ADJ
ijassa-348	8	19	,	,	PUNCT
ijassa-348	8	20	linear	linear	ADJ
ijassa-348	8	21	discriminant	discriminant	NOUN
ijassa-348	8	22	and	and	CCONJ
ijassa-348	8	23	quadratic	quadratic	ADJ
ijassa-348	8	24	discriminant	discriminant	NOUN
ijassa-348	8	25	analysis.the	analysis.the	DET
ijassa-348	8	26	aim	aim	NOUN
ijassa-348	8	27	of	of	ADP
ijassa-348	8	28	this	this	DET
ijassa-348	8	29	paper	paper	NOUN
ijassa-348	8	30	is	be	AUX
ijassa-348	8	31	to	to	PART
ijassa-348	8	32	identify	identify	VERB
ijassa-348	8	33	the	the	DET
ijassa-348	8	34	class	class	NOUN
ijassa-348	8	35	label	label	NOUN
ijassa-348	8	36	by	by	ADP
ijassa-348	8	37	constructing	construct	VERB
ijassa-348	8	38	an	an	DET
ijassa-348	8	39	individual	individual	ADJ
ijassa-348	8	40	classifier	classifier	NOUN
ijassa-348	8	41	for	for	ADP
ijassa-348	8	42	each	each	PRON
ijassa-348	8	43	of	of	ADP
ijassa-348	8	44	the	the	DET
ijassa-348	8	45	attack	attack	NOUN
ijassa-348	8	46	type	type	NOUN
ijassa-348	8	47	and	and	CCONJ
ijassa-348	8	48	merging	merge	VERB
ijassa-348	8	49	the	the	DET
ijassa-348	8	50	results	result	NOUN
ijassa-348	8	51	of	of	ADP
ijassa-348	8	52	every	every	DET
ijassa-348	8	53	classifier	classifier	NOUN
ijassa-348	8	54	.	.	PUNCT
ijassa-348	9	1	the	the	DET
ijassa-348	9	2	resultant	resultant	NOUN
ijassa-348	9	3	decision	decision	NOUN
ijassa-348	9	4	of	of	ADP
ijassa-348	9	5	the	the	DET
ijassa-348	9	6	class	class	NOUN
ijassa-348	9	7	label	label	NOUN
ijassa-348	9	8	is	be	AUX
ijassa-348	9	9	obtained	obtain	VERB
ijassa-348	9	10	using	use	VERB
ijassa-348	9	11	weighted	weight	VERB
ijassa-348	9	12	majority	majority	NOUN
ijassa-348	9	13	voting	voting	NOUN
ijassa-348	9	14	approach	approach	NOUN
ijassa-348	9	15	.	.	PUNCT
ijassa-348	10	1	we	we	PRON
ijassa-348	10	2	analyzed	analyze	VERB
ijassa-348	10	3	the	the	DET
ijassa-348	10	4	performance	performance	NOUN
ijassa-348	10	5	of	of	ADP
ijassa-348	10	6	the	the	DET
ijassa-348	10	7	model	model	NOUN
ijassa-348	10	8	on	on	ADP
ijassa-348	10	9	two	two	NUM
ijassa-348	10	10	different	different	ADJ
ijassa-348	10	11	data	datum	NOUN
ijassa-348	10	12	sets	set	NOUN
ijassa-348	10	13	such	such	ADJ
ijassa-348	10	14	as	as	ADP
ijassa-348	10	15	nsl	nsl	NOUN
ijassa-348	10	16	-	-	PUNCT
ijassa-348	10	17	kdd	kdd	PROPN
ijassa-348	10	18	and	and	CCONJ
ijassa-348	10	19	unsw	unsw	PROPN
ijassa-348	10	20	-	-	PUNCT
ijassa-348	10	21	nb	nb	NOUN
ijassa-348	10	22	datasets	dataset	NOUN
ijassa-348	10	23	.	.	PUNCT
ijassa-348	11	1	the	the	DET
ijassa-348	11	2	experimental	experimental	ADJ
ijassa-348	11	3	results	result	NOUN
ijassa-348	11	4	indicate	indicate	VERB
ijassa-348	11	5	that	that	SCONJ
ijassa-348	11	6	the	the	DET
ijassa-348	11	7	ensemble	ensemble	ADJ
ijassa-348	11	8	produces	produce	VERB
ijassa-348	11	9	high	high	ADJ
ijassa-348	11	10	accuracy	accuracy	NOUN
ijassa-348	11	11	in	in	ADP
ijassa-348	11	12	comparison	comparison	NOUN
ijassa-348	11	13	to	to	ADP
ijassa-348	11	14	the	the	DET
ijassa-348	11	15	base	base	NOUN
ijassa-348	11	16	classifiers	classifier	NOUN
ijassa-348	11	17	.	.	PUNCT
ijassa-348	12	1	as	as	SCONJ
ijassa-348	12	2	there	there	PRON
ijassa-348	12	3	is	be	VERB
ijassa-348	12	4	a	a	DET
ijassa-348	12	5	huge	huge	ADJ
ijassa-348	12	6	class	class	NOUN
ijassa-348	12	7	imbalance	imbalance	NOUN
ijassa-348	12	8	problem	problem	NOUN
ijassa-348	12	9	in	in	ADP
ijassa-348	12	10	network	network	NOUN
ijassa-348	12	11	traffic	traffic	NOUN
ijassa-348	12	12	,	,	PUNCT
ijassa-348	12	13	it	it	PRON
ijassa-348	12	14	is	be	AUX
ijassa-348	12	15	also	also	ADV
ijassa-348	12	16	observed	observe	VERB
ijassa-348	12	17	that	that	SCONJ
ijassa-348	12	18	rather	rather	ADV
ijassa-348	12	19	than	than	ADP
ijassa-348	12	20	relying	rely	VERB
ijassa-348	12	21	on	on	ADP
ijassa-348	12	22	a	a	DET
ijassa-348	12	23	single	single	ADJ
ijassa-348	12	24	classifier	classifier	NOUN
ijassa-348	12	25	,	,	PUNCT
ijassa-348	12	26	predicting	predict	VERB
ijassa-348	12	27	the	the	DET
ijassa-348	12	28	class	class	NOUN
ijassa-348	12	29	label	label	NOUN
ijassa-348	12	30	by	by	ADP
ijassa-348	12	31	weighted	weight	VERB
ijassa-348	12	32	majority	majority	NOUN
ijassa-348	12	33	voting	voting	NOUN
ijassa-348	12	34	of	of	ADP
ijassa-348	12	35	svm	svm	ADJ
ijassa-348	12	36	,	,	PUNCT
ijassa-348	12	37	linear	linear	ADJ
ijassa-348	12	38	and	and	CCONJ
ijassa-348	12	39	quadratic	quadratic	ADJ
ijassa-348	12	40	discriminant	discriminant	NOUN
ijassa-348	12	41	classifier	classifier	NOUN
ijassa-348	12	42	is	be	AUX
ijassa-348	12	43	an	an	DET
ijassa-348	12	44	optimal	optimal	ADJ
ijassa-348	12	45	solution	solution	NOUN
ijassa-348	12	46	which	which	PRON
ijassa-348	12	47	is	be	AUX
ijassa-348	12	48	proposed	propose	VERB
ijassa-348	12	49	in	in	ADP
ijassa-348	12	50	this	this	DET
ijassa-348	12	51	paper	paper	NOUN
ijassa-348	12	52	.	.	PUNCT
ijassa-348	13	1	keywords	keyword	NOUN
ijassa-348	13	2	accuracy	accuracy	NOUN
ijassa-348	13	3	;	;	PUNCT
ijassa-348	13	4	intrusion	intrusion	NOUN
ijassa-348	13	5	;	;	PUNCT
ijassa-348	13	6	linear	linear	ADJ
ijassa-348	13	7	discriminant	discriminant	NOUN
ijassa-348	13	8	;	;	PUNCT
ijassa-348	13	9	quadratic	quadratic	ADJ
ijassa-348	13	10	discriminant	discriminant	NOUN
ijassa-348	13	11	;	;	PUNCT
ijassa-348	13	12	support	support	NOUN
ijassa-348	13	13	vector	vector	NOUN
ijassa-348	13	14	machine	machine	NOUN
ijassa-348	13	15	.	.	PUNCT
ijassa-348	14	1	1	1	NUM
ijassa-348	14	2	introduction	introduction	NOUN
ijassa-348	14	3	malicious	malicious	ADJ
ijassa-348	14	4	intruders	intruder	NOUN
ijassa-348	14	5	in	in	ADP
ijassa-348	14	6	the	the	DET
ijassa-348	14	7	network	network	NOUN
ijassa-348	14	8	are	be	AUX
ijassa-348	14	9	increasing	increase	VERB
ijassa-348	14	10	day	day	NOUN
ijassa-348	14	11	by	by	ADP
ijassa-348	14	12	day	day	NOUN
ijassa-348	14	13	due	due	ADP
ijassa-348	14	14	to	to	ADP
ijassa-348	14	15	the	the	DET
ijassa-348	14	16	rapid	rapid	ADJ
ijassa-348	14	17	development	development	NOUN
ijassa-348	14	18	of	of	ADP
ijassa-348	14	19	internet	internet	NOUN
ijassa-348	14	20	.	.	PUNCT
ijassa-348	15	1	the	the	DET
ijassa-348	15	2	intruders	intruder	NOUN
ijassa-348	15	3	can	can	AUX
ijassa-348	15	4	access	access	VERB
ijassa-348	15	5	,	,	PUNCT
ijassa-348	15	6	manipulate	manipulate	VERB
ijassa-348	15	7	and	and	CCONJ
ijassa-348	15	8	disable	disable	VERB
ijassa-348	15	9	the	the	DET
ijassa-348	15	10	systems	system	NOUN
ijassa-348	15	11	connected	connect	VERB
ijassa-348	15	12	on	on	ADP
ijassa-348	15	13	the	the	DET
ijassa-348	15	14	internet	internet	NOUN
ijassa-348	15	15	.	.	PUNCT
ijassa-348	16	1	intrusion	intrusion	NOUN
ijassa-348	16	2	detection	detection	NOUN
ijassa-348	16	3	systems	system	NOUN
ijassa-348	16	4	(	(	PUNCT
ijassa-348	16	5	ids	id	NOUN
ijassa-348	16	6	)	)	PUNCT
ijassa-348	16	7	are	be	AUX
ijassa-348	16	8	designed	design	VERB
ijassa-348	16	9	to	to	PART
ijassa-348	16	10	discover	discover	VERB
ijassa-348	16	11	the	the	DET
ijassa-348	16	12	unauthorized	unauthorized	ADJ
ijassa-348	16	13	access	access	NOUN
ijassa-348	16	14	to	to	ADP
ijassa-348	16	15	computers	computer	NOUN
ijassa-348	16	16	in	in	ADP
ijassa-348	16	17	the	the	DET
ijassa-348	16	18	network	network	NOUN
ijassa-348	16	19	.	.	PUNCT
ijassa-348	17	1	intrusion	intrusion	NOUN
ijassa-348	17	2	detection	detection	NOUN
ijassa-348	17	3	systems	system	NOUN
ijassa-348	17	4	are	be	AUX
ijassa-348	17	5	classified	classify	VERB
ijassa-348	17	6	in	in	ADP
ijassa-348	17	7	two	two	NUM
ijassa-348	17	8	categories	category	NOUN
ijassa-348	17	9	signature	signature	NOUN
ijassa-348	17	10	detection	detection	NOUN
ijassa-348	17	11	and	and	CCONJ
ijassa-348	17	12	anomaly	anomaly	NOUN
ijassa-348	17	13	detection	detection	NOUN
ijassa-348	17	14	.	.	PUNCT
ijassa-348	18	1	signature	signature	NOUN
ijassa-348	18	2	detection	detection	NOUN
ijassa-348	18	3	is	be	AUX
ijassa-348	18	4	used	use	VERB
ijassa-348	18	5	to	to	PART
ijassa-348	18	6	identify	identify	VERB
ijassa-348	18	7	attacks	attack	NOUN
ijassa-348	18	8	based	base	VERB
ijassa-348	18	9	on	on	ADP
ijassa-348	18	10	the	the	DET
ijassa-348	18	11	known	know	VERB
ijassa-348	18	12	pattern	pattern	NOUN
ijassa-348	18	13	of	of	ADP
ijassa-348	18	14	attacks	attack	NOUN
ijassa-348	18	15	.	.	PUNCT
ijassa-348	19	1	anomaly	anomaly	PROPN
ijassa-348	19	2	detection	detection	NOUN
ijassa-348	19	3	compares	compare	VERB
ijassa-348	19	4	unknown	unknown	ADJ
ijassa-348	19	5	profiles	profile	NOUN
ijassa-348	19	6	with	with	ADP
ijassa-348	19	7	known	know	VERB
ijassa-348	19	8	profiles	profile	NOUN
ijassa-348	19	9	and	and	CCONJ
ijassa-348	19	10	then	then	ADV
ijassa-348	19	11	identifies	identify	VERB
ijassa-348	19	12	the	the	DET
ijassa-348	19	13	unknown	unknown	ADJ
ijassa-348	19	14	traffic	traffic	NOUN
ijassa-348	19	15	profile	profile	NOUN
ijassa-348	19	16	as	as	ADP
ijassa-348	19	17	an	an	DET
ijassa-348	19	18	attack	attack	NOUN
ijassa-348	19	19	.	.	PUNCT
ijassa-348	20	1	anomaly	anomaly	PROPN
ijassa-348	20	2	detection	detection	NOUN
ijassa-348	20	3	techniques	technique	NOUN
ijassa-348	20	4	have	have	VERB
ijassa-348	20	5	high	high	ADJ
ijassa-348	20	6	false	false	ADJ
ijassa-348	20	7	positive	positive	ADJ
ijassa-348	20	8	rates	rate	NOUN
ijassa-348	20	9	.	.	PUNCT
ijassa-348	21	1	many	many	ADJ
ijassa-348	21	2	machine	machine	NOUN
ijassa-348	21	3	learning	learn	VERB
ijassa-348	21	4	techniques	technique	NOUN
ijassa-348	21	5	have	have	AUX
ijassa-348	21	6	been	be	AUX
ijassa-348	21	7	used	use	VERB
ijassa-348	21	8	by	by	ADP
ijassa-348	21	9	researchers	researcher	NOUN
ijassa-348	21	10	to	to	PART
ijassa-348	21	11	overcome	overcome	VERB
ijassa-348	21	12	the	the	DET
ijassa-348	21	13	disadvantages	disadvantage	NOUN
ijassa-348	21	14	of	of	ADP
ijassa-348	21	15	anomaly	anomaly	NOUN
ijassa-348	21	16	detection	detection	NOUN
ijassa-348	21	17	models	model	NOUN
ijassa-348	21	18	.	.	PUNCT
ijassa-348	22	1	several	several	ADJ
ijassa-348	22	2	intelligent	intelligent	ADJ
ijassa-348	22	3	approaches	approach	NOUN
ijassa-348	22	4	such	such	ADJ
ijassa-348	22	5	as	as	ADP
ijassa-348	22	6	svms	svms	NOUN
ijassa-348	22	7	[	[	X
ijassa-348	22	8	1	1	NUM
ijassa-348	22	9	]	]	PUNCT
ijassa-348	22	10	,	,	PUNCT
ijassa-348	22	11	16	16	NUM
ijassa-348	22	12	sumaiya	sumaiya	NOUN
ijassa-348	22	13	thaseen	thaseen	PROPN
ijassa-348	22	14	and	and	CCONJ
ijassa-348	22	15	ch.aswani	ch.aswani	X
ijassa-348	23	1	kumar	kumar	PROPN
ijassa-348	23	2	:	:	PUNCT
ijassa-348	23	3	intrusion	intrusion	NOUN
ijassa-348	23	4	detection	detection	NOUN
ijassa-348	23	5	model	model	NOUN
ijassa-348	23	6	using	use	VERB
ijassa-348	23	7	pca	pca	PROPN
ijassa-348	23	8	and	and	CCONJ
ijassa-348	23	9	...	...	PUNCT
ijassa-348	23	10	anns	anns	PROPN
ijassa-348	23	11	[	[	X
ijassa-348	23	12	2	2	NUM
ijassa-348	23	13	]	]	PUNCT
ijassa-348	23	14	,	,	PUNCT
ijassa-348	23	15	petri	petri	ADJ
ijassa-348	23	16	nets	net	NOUN
ijassa-348	23	17	and	and	CCONJ
ijassa-348	23	18	data	datum	NOUN
ijassa-348	23	19	mining	mining	NOUN
ijassa-348	23	20	approaches[3	approaches[3	PROPN
ijassa-348	23	21	,	,	PUNCT
ijassa-348	23	22	4	4	NUM
ijassa-348	23	23	]	]	PUNCT
ijassa-348	23	24	have	have	AUX
ijassa-348	23	25	been	be	AUX
ijassa-348	23	26	used	use	VERB
ijassa-348	23	27	to	to	PART
ijassa-348	23	28	build	build	VERB
ijassa-348	23	29	an	an	DET
ijassa-348	23	30	ids	id	NOUN
ijassa-348	23	31	.	.	PUNCT
ijassa-348	24	1	there	there	PRON
ijassa-348	24	2	are	be	VERB
ijassa-348	24	3	also	also	ADV
ijassa-348	24	4	many	many	ADJ
ijassa-348	24	5	ensemble	ensemble	ADJ
ijassa-348	24	6	methods	method	NOUN
ijassa-348	24	7	of	of	ADP
ijassa-348	24	8	machine	machine	NOUN
ijassa-348	24	9	learning	learning	NOUN
ijassa-348	24	10	which	which	PRON
ijassa-348	24	11	are	be	AUX
ijassa-348	24	12	more	more	ADV
ijassa-348	24	13	efficient	efficient	ADJ
ijassa-348	24	14	than	than	ADP
ijassa-348	24	15	individual	individual	ADJ
ijassa-348	24	16	techniques	technique	NOUN
ijassa-348	24	17	that	that	PRON
ijassa-348	24	18	can	can	AUX
ijassa-348	24	19	reduce	reduce	VERB
ijassa-348	24	20	the	the	DET
ijassa-348	24	21	false	false	ADJ
ijassa-348	24	22	alarm	alarm	NOUN
ijassa-348	24	23	rate	rate	NOUN
ijassa-348	24	24	and	and	CCONJ
ijassa-348	24	25	increase	increase	VERB
ijassa-348	24	26	the	the	DET
ijassa-348	24	27	classification	classification	NOUN
ijassa-348	24	28	accuracy	accuracy	NOUN
ijassa-348	24	29	.	.	PUNCT
ijassa-348	25	1	the	the	DET
ijassa-348	25	2	different	different	ADJ
ijassa-348	25	3	ensemble	ensemble	ADJ
ijassa-348	25	4	methods	method	NOUN
ijassa-348	25	5	are	be	AUX
ijassa-348	25	6	bagging	bag	VERB
ijassa-348	25	7	,	,	PUNCT
ijassa-348	25	8	boosting	boost	VERB
ijassa-348	25	9	and	and	CCONJ
ijassa-348	25	10	stacking	stacking	NOUN
ijassa-348	25	11	.	.	PUNCT
ijassa-348	26	1	bagging	bagging	NOUN
ijassa-348	26	2	and	and	CCONJ
ijassa-348	26	3	boosting	boost	VERB
ijassa-348	26	4	are	be	AUX
ijassa-348	26	5	mostly	mostly	ADV
ijassa-348	26	6	used	use	VERB
ijassa-348	26	7	to	to	PART
ijassa-348	26	8	implement	implement	VERB
ijassa-348	26	9	intrusion	intrusion	NOUN
ijassa-348	26	10	detection	detection	NOUN
ijassa-348	26	11	models	model	NOUN
ijassa-348	26	12	as	as	SCONJ
ijassa-348	26	13	the	the	DET
ijassa-348	26	14	stacking	stacking	NOUN
ijassa-348	26	15	technique	technique	NOUN
ijassa-348	26	16	requires	require	VERB
ijassa-348	26	17	more	more	ADJ
ijassa-348	26	18	time	time	NOUN
ijassa-348	26	19	.	.	PUNCT
ijassa-348	27	1	there	there	PRON
ijassa-348	27	2	are	be	VERB
ijassa-348	27	3	two	two	NUM
ijassa-348	27	4	major	major	ADJ
ijassa-348	27	5	limitations	limitation	NOUN
ijassa-348	27	6	of	of	ADP
ijassa-348	27	7	existing	exist	VERB
ijassa-348	27	8	approaches	approach	NOUN
ijassa-348	27	9	.	.	PUNCT
ijassa-348	28	1	the	the	DET
ijassa-348	28	2	first	first	ADJ
ijassa-348	28	3	limitation	limitation	NOUN
ijassa-348	28	4	is	be	AUX
ijassa-348	28	5	even	even	ADV
ijassa-348	28	6	though	though	SCONJ
ijassa-348	28	7	there	there	PRON
ijassa-348	28	8	are	be	VERB
ijassa-348	28	9	many	many	ADJ
ijassa-348	28	10	sophisticated	sophisticated	ADJ
ijassa-348	28	11	detection	detection	NOUN
ijassa-348	28	12	techniques	technique	NOUN
ijassa-348	28	13	only	only	ADV
ijassa-348	28	14	few	few	ADJ
ijassa-348	28	15	focus	focus	NOUN
ijassa-348	28	16	on	on	ADP
ijassa-348	28	17	feature	feature	NOUN
ijassa-348	28	18	representation	representation	NOUN
ijassa-348	28	19	for	for	ADP
ijassa-348	28	20	normal	normal	ADJ
ijassa-348	28	21	traffic	traffic	NOUN
ijassa-348	28	22	and	and	CCONJ
ijassa-348	28	23	attack	attack	NOUN
ijassa-348	28	24	traffic	traffic	NOUN
ijassa-348	28	25	which	which	PRON
ijassa-348	28	26	is	be	AUX
ijassa-348	28	27	a	a	DET
ijassa-348	28	28	major	major	ADJ
ijassa-348	28	29	issue	issue	NOUN
ijassa-348	28	30	to	to	PART
ijassa-348	28	31	enhance	enhance	VERB
ijassa-348	28	32	performance	performance	NOUN
ijassa-348	28	33	of	of	ADP
ijassa-348	28	34	the	the	DET
ijassa-348	28	35	classifier	classifier	NOUN
ijassa-348	28	36	.	.	PUNCT
ijassa-348	29	1	the	the	DET
ijassa-348	29	2	second	second	ADJ
ijassa-348	29	3	issue	issue	NOUN
ijassa-348	29	4	is	be	AUX
ijassa-348	29	5	the	the	DET
ijassa-348	29	6	computation	computation	NOUN
ijassa-348	29	7	time	time	NOUN
ijassa-348	29	8	involved	involve	VERB
ijassa-348	29	9	in	in	ADP
ijassa-348	29	10	integrating	integrate	VERB
ijassa-348	29	11	multiple	multiple	ADJ
ijassa-348	29	12	techniques	technique	NOUN
ijassa-348	29	13	which	which	PRON
ijassa-348	29	14	may	may	AUX
ijassa-348	29	15	degrade	degrade	VERB
ijassa-348	29	16	the	the	DET
ijassa-348	29	17	efficiency	efficiency	NOUN
ijassa-348	29	18	of	of	ADP
ijassa-348	29	19	on	on	ADP
ijassa-348	29	20	-	-	PUNCT
ijassa-348	29	21	line	line	NOUN
ijassa-348	29	22	detection	detection	NOUN
ijassa-348	29	23	.	.	PUNCT
ijassa-348	30	1	the	the	DET
ijassa-348	30	2	contribution	contribution	NOUN
ijassa-348	30	3	of	of	ADP
ijassa-348	30	4	this	this	DET
ijassa-348	30	5	research	research	NOUN
ijassa-348	30	6	is	be	AUX
ijassa-348	30	7	to	to	PART
ijassa-348	30	8	to	to	PART
ijassa-348	30	9	build	build	VERB
ijassa-348	30	10	a	a	DET
ijassa-348	30	11	desirable	desirable	ADJ
ijassa-348	30	12	ids	id	NOUN
ijassa-348	30	13	model	model	NOUN
ijassa-348	30	14	with	with	ADP
ijassa-348	30	15	high	high	ADJ
ijassa-348	30	16	accuracy	accuracy	NOUN
ijassa-348	30	17	using	use	VERB
ijassa-348	30	18	machine	machine	NOUN
ijassa-348	30	19	learning	learn	VERB
ijassa-348	30	20	ensemble	ensemble	ADJ
ijassa-348	30	21	techniques	technique	NOUN
ijassa-348	30	22	with	with	ADP
ijassa-348	30	23	a	a	DET
ijassa-348	30	24	feature	feature	NOUN
ijassa-348	30	25	reduction	reduction	NOUN
ijassa-348	30	26	technique	technique	NOUN
ijassa-348	30	27	to	to	PART
ijassa-348	30	28	identify	identify	VERB
ijassa-348	30	29	the	the	DET
ijassa-348	30	30	suitable	suitable	ADJ
ijassa-348	30	31	features	feature	NOUN
ijassa-348	30	32	in	in	ADP
ijassa-348	30	33	ids	id	NOUN
ijassa-348	30	34	.	.	PUNCT
ijassa-348	31	1	ensemble	ensemble	ADJ
ijassa-348	31	2	is	be	AUX
ijassa-348	31	3	preferred	prefer	VERB
ijassa-348	31	4	because	because	SCONJ
ijassa-348	31	5	the	the	DET
ijassa-348	31	6	aggregation	aggregation	NOUN
ijassa-348	31	7	of	of	ADP
ijassa-348	31	8	multiple	multiple	ADJ
ijassa-348	31	9	classifier	classifier	NOUN
ijassa-348	31	10	predictions	prediction	NOUN
ijassa-348	31	11	improves	improve	VERB
ijassa-348	31	12	the	the	DET
ijassa-348	31	13	accuracy	accuracy	NOUN
ijassa-348	31	14	of	of	ADP
ijassa-348	31	15	ids	id	NOUN
ijassa-348	31	16	.	.	PUNCT
ijassa-348	32	1	in	in	ADP
ijassa-348	32	2	this	this	DET
ijassa-348	32	3	paper	paper	NOUN
ijassa-348	32	4	we	we	PRON
ijassa-348	32	5	specify	specify	VERB
ijassa-348	32	6	a	a	DET
ijassa-348	32	7	manager	manager	NOUN
ijassa-348	32	8	as	as	ADP
ijassa-348	32	9	a	a	DET
ijassa-348	32	10	combination	combination	NOUN
ijassa-348	32	11	of	of	ADP
ijassa-348	32	12	classifiers	classifier	NOUN
ijassa-348	32	13	that	that	PRON
ijassa-348	32	14	will	will	AUX
ijassa-348	32	15	be	be	AUX
ijassa-348	32	16	generated	generate	VERB
ijassa-348	32	17	depending	depend	VERB
ijassa-348	32	18	on	on	ADP
ijassa-348	32	19	the	the	DET
ijassa-348	32	20	class	class	NOUN
ijassa-348	32	21	labels	label	NOUN
ijassa-348	32	22	in	in	ADP
ijassa-348	32	23	each	each	DET
ijassa-348	32	24	dataset	dataset	NOUN
ijassa-348	32	25	.	.	PUNCT
ijassa-348	33	1	for	for	ADP
ijassa-348	33	2	instance	instance	NOUN
ijassa-348	33	3	,	,	PUNCT
ijassa-348	33	4	the	the	DET
ijassa-348	33	5	manager	manager	NOUN
ijassa-348	33	6	will	will	AUX
ijassa-348	33	7	have	have	VERB
ijassa-348	33	8	5	5	NUM
ijassa-348	33	9	different	different	ADJ
ijassa-348	33	10	classifiers	classifier	NOUN
ijassa-348	33	11	in	in	ADP
ijassa-348	33	12	the	the	DET
ijassa-348	33	13	svm	svm	PROPN
ijassa-348	33	14	,	,	PUNCT
ijassa-348	33	15	mnb	mnb	PROPN
ijassa-348	33	16	and	and	CCONJ
ijassa-348	33	17	ldc	ldc	PROPN
ijassa-348	33	18	ensemble	ensemble	ADJ
ijassa-348	33	19	for	for	ADP
ijassa-348	33	20	nsl	nsl	PROPN
ijassa-348	33	21	-	-	PUNCT
ijassa-348	33	22	kdd	kdd	PROPN
ijassa-348	33	23	data	datum	NOUN
ijassa-348	33	24	set	set	VERB
ijassa-348	33	25	as	as	SCONJ
ijassa-348	33	26	there	there	PRON
ijassa-348	33	27	are	be	VERB
ijassa-348	33	28	five	five	NUM
ijassa-348	33	29	class	class	NOUN
ijassa-348	33	30	labels	label	NOUN
ijassa-348	33	31	and	and	CCONJ
ijassa-348	33	32	the	the	DET
ijassa-348	33	33	manager	manager	NOUN
ijassa-348	33	34	will	will	AUX
ijassa-348	33	35	have	have	VERB
ijassa-348	33	36	9	9	NUM
ijassa-348	33	37	different	different	ADJ
ijassa-348	33	38	classifiers	classifier	NOUN
ijassa-348	33	39	in	in	ADP
ijassa-348	33	40	the	the	DET
ijassa-348	33	41	svm	svm	PROPN
ijassa-348	33	42	,	,	PUNCT
ijassa-348	33	43	ldc	ldc	PROPN
ijassa-348	33	44	and	and	CCONJ
ijassa-348	33	45	qdc	qdc	VERB
ijassa-348	33	46	for	for	ADP
ijassa-348	33	47	unswnb	unswnb	ADJ
ijassa-348	33	48	dataset	dataset	NOUN
ijassa-348	33	49	.	.	PUNCT
ijassa-348	34	1	thus	thus	ADV
ijassa-348	34	2	the	the	DET
ijassa-348	34	3	manager	manager	NOUN
ijassa-348	34	4	has	have	VERB
ijassa-348	34	5	a	a	DET
ijassa-348	34	6	variable	variable	ADJ
ijassa-348	34	7	number	number	NOUN
ijassa-348	34	8	of	of	ADP
ijassa-348	34	9	classifiers	classifier	NOUN
ijassa-348	34	10	generated	generate	VERB
ijassa-348	34	11	dynamically	dynamically	ADV
ijassa-348	34	12	according	accord	VERB
ijassa-348	34	13	to	to	ADP
ijassa-348	34	14	the	the	DET
ijassa-348	34	15	number	number	NOUN
ijassa-348	34	16	of	of	ADP
ijassa-348	34	17	class	class	NOUN
ijassa-348	34	18	labels	label	NOUN
ijassa-348	34	19	available	available	ADJ
ijassa-348	34	20	in	in	ADP
ijassa-348	34	21	the	the	DET
ijassa-348	34	22	dataset	dataset	NOUN
ijassa-348	34	23	.	.	PUNCT
ijassa-348	35	1	thus	thus	ADV
ijassa-348	35	2	the	the	DET
ijassa-348	35	3	ensemble	ensemble	ADJ
ijassa-348	35	4	model	model	NOUN
ijassa-348	35	5	utilizes	utilize	VERB
ijassa-348	35	6	a	a	DET
ijassa-348	35	7	individual	individual	ADJ
ijassa-348	35	8	classifier	classifier	NOUN
ijassa-348	35	9	for	for	ADP
ijassa-348	35	10	every	every	DET
ijassa-348	35	11	class	class	NOUN
ijassa-348	35	12	type	type	NOUN
ijassa-348	35	13	and	and	CCONJ
ijassa-348	35	14	is	be	AUX
ijassa-348	35	15	an	an	DET
ijassa-348	35	16	integration	integration	NOUN
ijassa-348	35	17	of	of	ADP
ijassa-348	35	18	base	base	NOUN
ijassa-348	35	19	classifiers	classifier	NOUN
ijassa-348	35	20	svm	svm	VERB
ijassa-348	35	21	,	,	PUNCT
ijassa-348	35	22	linear	linear	ADJ
ijassa-348	35	23	discriminant	discriminant	NOUN
ijassa-348	35	24	and	and	CCONJ
ijassa-348	35	25	quadratic	quadratic	ADJ
ijassa-348	35	26	discriminant	discriminant	NOUN
ijassa-348	35	27	with	with	ADP
ijassa-348	35	28	the	the	DET
ijassa-348	35	29	resultant	resultant	NOUN
ijassa-348	35	30	class	class	NOUN
ijassa-348	35	31	label	label	NOUN
ijassa-348	35	32	predicted	predict	VERB
ijassa-348	35	33	by	by	ADP
ijassa-348	35	34	weighted	weight	VERB
ijassa-348	35	35	majority	majority	NOUN
ijassa-348	35	36	voting	voting	NOUN
ijassa-348	35	37	ensemble	ensemble	ADJ
ijassa-348	35	38	which	which	PRON
ijassa-348	35	39	is	be	AUX
ijassa-348	35	40	deployed	deploy	VERB
ijassa-348	35	41	to	to	PART
ijassa-348	35	42	classify	classify	VERB
ijassa-348	35	43	the	the	DET
ijassa-348	35	44	different	different	ADJ
ijassa-348	35	45	kinds	kind	NOUN
ijassa-348	35	46	of	of	ADP
ijassa-348	35	47	network	network	NOUN
ijassa-348	35	48	attacks	attack	NOUN
ijassa-348	35	49	.	.	PUNCT
ijassa-348	36	1	a	a	DET
ijassa-348	36	2	similar	similar	ADJ
ijassa-348	36	3	ensemble	ensemble	NOUN
ijassa-348	36	4	of	of	ADP
ijassa-348	36	5	classifiers	classifier	NOUN
ijassa-348	36	6	was	be	AUX
ijassa-348	36	7	already	already	ADV
ijassa-348	36	8	developed	develop	VERB
ijassa-348	36	9	[	[	X
ijassa-348	36	10	5	5	NUM
ijassa-348	36	11	]	]	PUNCT
ijassa-348	36	12	using	use	VERB
ijassa-348	36	13	four	four	NUM
ijassa-348	36	14	dif	dif	ADJ
ijassa-348	36	15	-	-	PUNCT
ijassa-348	36	16	ferent	ferent	NOUN
ijassa-348	36	17	base	base	NOUN
ijassa-348	36	18	classifiers	classifier	NOUN
ijassa-348	36	19	namely	namely	ADV
ijassa-348	36	20	linear	linear	VERB
ijassa-348	36	21	discriminant	discriminant	NOUN
ijassa-348	36	22	classifier	classifier	NOUN
ijassa-348	36	23	(	(	PUNCT
ijassa-348	36	24	ldc	ldc	PROPN
ijassa-348	36	25	)	)	PUNCT
ijassa-348	36	26	,	,	PUNCT
ijassa-348	36	27	quadratic	quadratic	ADJ
ijassa-348	36	28	discriminant	discriminant	NOUN
ijassa-348	36	29	classifier	classifier	NOUN
ijassa-348	36	30	(	(	PUNCT
ijassa-348	36	31	qdc	qdc	PROPN
ijassa-348	36	32	)	)	PUNCT
ijassa-348	36	33	,	,	PUNCT
ijassa-348	36	34	k	k	X
ijassa-348	36	35	-	-	PUNCT
ijassa-348	36	36	nearest	near	ADJ
ijassa-348	36	37	neighbor	neighbor	NOUN
ijassa-348	36	38	(	(	PUNCT
ijassa-348	36	39	knn	knn	PROPN
ijassa-348	36	40	)	)	PUNCT
ijassa-348	36	41	and	and	CCONJ
ijassa-348	36	42	back	back	ADJ
ijassa-348	36	43	propagation	propagation	NOUN
ijassa-348	36	44	and	and	CCONJ
ijassa-348	36	45	tested	test	VERB
ijassa-348	36	46	on	on	ADP
ijassa-348	36	47	four	four	NUM
ijassa-348	36	48	datasets	dataset	NOUN
ijassa-348	36	49	namely	namely	ADV
ijassa-348	36	50	hearth	hearth	NOUN
ijassa-348	36	51	,	,	PUNCT
ijassa-348	36	52	diabetes	diabetes	NOUN
ijassa-348	36	53	,	,	PUNCT
ijassa-348	36	54	iris	iris	NOUN
ijassa-348	36	55	and	and	CCONJ
ijassa-348	36	56	transfusion	transfusion	NOUN
ijassa-348	36	57	.	.	PUNCT
ijassa-348	37	1	the	the	DET
ijassa-348	37	2	uniqueness	uniqueness	NOUN
ijassa-348	37	3	of	of	ADP
ijassa-348	37	4	the	the	DET
ijassa-348	37	5	proposed	propose	VERB
ijassa-348	37	6	model	model	NOUN
ijassa-348	37	7	over	over	ADP
ijassa-348	37	8	earlier	early	ADV
ijassa-348	37	9	developed	develop	VERB
ijassa-348	37	10	ensemble	ensemble	ADJ
ijassa-348	37	11	techniques	technique	NOUN
ijassa-348	37	12	are	be	AUX
ijassa-348	37	13	1	1	NUM
ijassa-348	37	14	)	)	PUNCT
ijassa-348	37	15	the	the	DET
ijassa-348	37	16	model	model	NOUN
ijassa-348	37	17	can	can	AUX
ijassa-348	37	18	be	be	AUX
ijassa-348	37	19	evaluated	evaluate	VERB
ijassa-348	37	20	on	on	ADP
ijassa-348	37	21	any	any	DET
ijassa-348	37	22	real	real	ADJ
ijassa-348	37	23	time	time	NOUN
ijassa-348	37	24	intrusion	intrusion	NOUN
ijassa-348	37	25	detection	detection	NOUN
ijassa-348	37	26	datasets	dataset	NOUN
ijassa-348	37	27	and	and	CCONJ
ijassa-348	37	28	the	the	DET
ijassa-348	37	29	managers	manager	NOUN
ijassa-348	37	30	can	can	AUX
ijassa-348	37	31	be	be	AUX
ijassa-348	37	32	extended	extend	VERB
ijassa-348	37	33	based	base	VERB
ijassa-348	37	34	on	on	ADP
ijassa-348	37	35	the	the	DET
ijassa-348	37	36	number	number	NOUN
ijassa-348	37	37	of	of	ADP
ijassa-348	37	38	class	class	NOUN
ijassa-348	37	39	labels	label	NOUN
ijassa-348	37	40	in	in	ADP
ijassa-348	37	41	the	the	DET
ijassa-348	37	42	samples	sample	NOUN
ijassa-348	37	43	.	.	PUNCT
ijassa-348	38	1	2	2	X
ijassa-348	38	2	)	)	PUNCT
ijassa-348	38	3	any	any	DET
ijassa-348	38	4	combination	combination	NOUN
ijassa-348	38	5	of	of	ADP
ijassa-348	38	6	base	base	NOUN
ijassa-348	38	7	classifiers	classifier	NOUN
ijassa-348	38	8	can	can	AUX
ijassa-348	38	9	replace	replace	VERB
ijassa-348	38	10	the	the	DET
ijassa-348	38	11	existing	exist	VERB
ijassa-348	38	12	techniques	technique	NOUN
ijassa-348	38	13	to	to	PART
ijassa-348	38	14	improve	improve	VERB
ijassa-348	38	15	the	the	DET
ijassa-348	38	16	model	model	NOUN
ijassa-348	38	17	.	.	PUNCT
ijassa-348	39	1	the	the	DET
ijassa-348	39	2	rest	rest	NOUN
ijassa-348	39	3	of	of	ADP
ijassa-348	39	4	this	this	DET
ijassa-348	39	5	paper	paper	NOUN
ijassa-348	39	6	is	be	AUX
ijassa-348	39	7	summarized	summarize	VERB
ijassa-348	39	8	as	as	ADP
ijassa-348	39	9	follows	follow	NOUN
ijassa-348	39	10	.	.	PUNCT
ijassa-348	40	1	section	section	NOUN
ijassa-348	40	2	2	2	NUM
ijassa-348	40	3	provides	provide	VERB
ijassa-348	40	4	a	a	DET
ijassa-348	40	5	discussion	discussion	NOUN
ijassa-348	40	6	on	on	ADP
ijassa-348	40	7	various	various	ADJ
ijassa-348	40	8	developed	develop	VERB
ijassa-348	40	9	intrusion	intrusion	NOUN
ijassa-348	40	10	detection	detection	NOUN
ijassa-348	40	11	models	model	NOUN
ijassa-348	40	12	.	.	PUNCT
ijassa-348	41	1	section	section	NOUN
ijassa-348	41	2	3	3	NUM
ijassa-348	41	3	provides	provide	VERB
ijassa-348	41	4	the	the	DET
ijassa-348	41	5	background	background	NOUN
ijassa-348	41	6	of	of	ADP
ijassa-348	41	7	various	various	ADJ
ijassa-348	41	8	techniques	technique	NOUN
ijassa-348	41	9	utilized	utilize	VERB
ijassa-348	41	10	in	in	ADP
ijassa-348	41	11	the	the	DET
ijassa-348	41	12	model	model	NOUN
ijassa-348	41	13	such	such	ADJ
ijassa-348	41	14	as	as	ADP
ijassa-348	41	15	svm	svm	ADJ
ijassa-348	41	16	,	,	PUNCT
ijassa-348	41	17	linear	linear	ADJ
ijassa-348	41	18	disadvances	disadvance	NOUN
ijassa-348	41	19	in	in	ADP
ijassa-348	41	20	systems	system	NOUN
ijassa-348	41	21	science	science	NOUN
ijassa-348	41	22	and	and	CCONJ
ijassa-348	41	23	application	application	NOUN
ijassa-348	41	24	(	(	PUNCT
ijassa-348	41	25	2016	2016	NUM
ijassa-348	41	26	)	)	PUNCT
ijassa-348	42	1	vol.16	vol.16	PROPN
ijassa-348	42	2	no.2	no.2	PROPN
ijassa-348	42	3	17	17	NUM
ijassa-348	42	4	criminant	criminant	NOUN
ijassa-348	42	5	,	,	PUNCT
ijassa-348	42	6	quadratic	quadratic	ADJ
ijassa-348	42	7	discriminant	discriminant	NOUN
ijassa-348	42	8	classifier	classifier	NOUN
ijassa-348	42	9	and	and	CCONJ
ijassa-348	42	10	weighted	weight	VERB
ijassa-348	42	11	majority	majority	NOUN
ijassa-348	42	12	approach	approach	NOUN
ijassa-348	42	13	.	.	PUNCT
ijassa-348	43	1	section	section	NOUN
ijassa-348	43	2	4	4	NUM
ijassa-348	43	3	gives	give	VERB
ijassa-348	43	4	the	the	DET
ijassa-348	43	5	overview	overview	NOUN
ijassa-348	43	6	of	of	ADP
ijassa-348	43	7	proposed	propose	VERB
ijassa-348	43	8	intrusion	intrusion	NOUN
ijassa-348	43	9	detection	detection	NOUN
ijassa-348	43	10	model	model	NOUN
ijassa-348	43	11	.	.	PUNCT
ijassa-348	44	1	experimental	experimental	ADJ
ijassa-348	44	2	results	result	NOUN
ijassa-348	44	3	and	and	CCONJ
ijassa-348	44	4	discussions	discussion	NOUN
ijassa-348	44	5	are	be	AUX
ijassa-348	44	6	discussed	discuss	VERB
ijassa-348	44	7	in	in	ADP
ijassa-348	44	8	section	section	NOUN
ijassa-348	44	9	5	5	NUM
ijassa-348	44	10	.	.	PUNCT
ijassa-348	44	11	section	section	NOUN
ijassa-348	44	12	6	6	NUM
ijassa-348	44	13	finally	finally	ADV
ijassa-348	44	14	concludes	conclude	VERB
ijassa-348	44	15	the	the	DET
ijassa-348	44	16	paper	paper	NOUN
ijassa-348	44	17	.	.	PUNCT
ijassa-348	45	1	2	2	NUM
ijassa-348	45	2	related	relate	VERB
ijassa-348	45	3	work	work	NOUN
ijassa-348	45	4	in	in	ADP
ijassa-348	45	5	this	this	DET
ijassa-348	45	6	section	section	NOUN
ijassa-348	45	7	we	we	PRON
ijassa-348	45	8	will	will	AUX
ijassa-348	45	9	discuss	discuss	VERB
ijassa-348	45	10	the	the	DET
ijassa-348	45	11	various	various	ADJ
ijassa-348	45	12	intrusion	intrusion	NOUN
ijassa-348	45	13	detection	detection	NOUN
ijassa-348	45	14	models	model	NOUN
ijassa-348	45	15	developed	develop	VERB
ijassa-348	45	16	using	use	VERB
ijassa-348	45	17	machine	machine	NOUN
ijassa-348	45	18	learning	learning	NOUN
ijassa-348	45	19	approaches	approach	NOUN
ijassa-348	45	20	,	,	PUNCT
ijassa-348	45	21	models	model	NOUN
ijassa-348	45	22	developed	develop	VERB
ijassa-348	45	23	by	by	ADP
ijassa-348	45	24	integrating	integrate	VERB
ijassa-348	45	25	classifiers	classifier	NOUN
ijassa-348	45	26	and	and	CCONJ
ijassa-348	45	27	the	the	DET
ijassa-348	45	28	ensemble	ensemble	ADJ
ijassa-348	45	29	techniques	technique	NOUN
ijassa-348	45	30	developed	develop	VERB
ijassa-348	45	31	for	for	ADP
ijassa-348	45	32	intrusion	intrusion	NOUN
ijassa-348	45	33	detection	detection	NOUN
ijassa-348	45	34	.	.	PUNCT
ijassa-348	46	1	various	various	ADJ
ijassa-348	46	2	artificial	artificial	ADJ
ijassa-348	46	3	intelligence	intelligence	NOUN
ijassa-348	46	4	methods	method	NOUN
ijassa-348	46	5	have	have	AUX
ijassa-348	46	6	been	be	AUX
ijassa-348	46	7	developed	develop	VERB
ijassa-348	46	8	for	for	ADP
ijassa-348	46	9	intrusion	intrusion	NOUN
ijassa-348	46	10	detection	detection	NOUN
ijassa-348	46	11	models	model	NOUN
ijassa-348	46	12	such	such	ADJ
ijassa-348	46	13	as	as	ADP
ijassa-348	46	14	fuzzy	fuzzy	ADJ
ijassa-348	46	15	logic	logic	NOUN
ijassa-348	46	16	[	[	X
ijassa-348	46	17	6	6	NUM
ijassa-348	46	18	]	]	PUNCT
ijassa-348	46	19	,	,	PUNCT
ijassa-348	46	20	k	k	X
ijassa-348	46	21	-	-	PUNCT
ijassa-348	46	22	nearest	near	ADJ
ijassa-348	46	23	neighbors	neighbor	NOUN
ijassa-348	46	24	[	[	X
ijassa-348	46	25	7	7	NUM
ijassa-348	46	26	]	]	PUNCT
ijassa-348	46	27	,	,	PUNCT
ijassa-348	46	28	support	support	VERB
ijassa-348	46	29	vector	vector	NOUN
ijassa-348	46	30	machines	machine	NOUN
ijassa-348	46	31	[	[	X
ijassa-348	46	32	1	1	NUM
ijassa-348	46	33	]	]	PUNCT
ijassa-348	46	34	,	,	PUNCT
ijassa-348	46	35	artificial	artificial	ADJ
ijassa-348	46	36	neural	neural	ADJ
ijassa-348	46	37	networks	network	NOUN
ijassa-348	47	1	[	[	X
ijassa-348	47	2	2	2	NUM
ijassa-348	47	3	]	]	PUNCT
ijassa-348	47	4	,	,	PUNCT
ijassa-348	47	5	naïve	naïve	ADJ
ijassa-348	47	6	bayes	bayes	PROPN
ijassa-348	47	7	networks	network	NOUN
ijassa-348	48	1	[	[	X
ijassa-348	48	2	8	8	NUM
ijassa-348	48	3	]	]	PUNCT
ijassa-348	48	4	,	,	PUNCT
ijassa-348	48	5	decision	decision	NOUN
ijassa-348	48	6	trees	tree	NOUN
ijassa-348	48	7	[	[	X
ijassa-348	48	8	9	9	NUM
ijassa-348	48	9	]	]	PUNCT
ijassa-348	48	10	,	,	PUNCT
ijassa-348	48	11	and	and	CCONJ
ijassa-348	48	12	genetic	genetic	ADJ
ijassa-348	48	13	algorithms	algorithm	NOUN
ijassa-348	48	14	[	[	X
ijassa-348	48	15	10	10	NUM
ijassa-348	48	16	]	]	PUNCT
ijassa-348	48	17	.	.	PUNCT
ijassa-348	49	1	sumaiya	sumaiya	PROPN
ijassa-348	49	2	et	et	PROPN
ijassa-348	49	3	al	al	PROPN
ijassa-348	49	4	.	.	PROPN
ijassa-348	49	5	developed	develop	VERB
ijassa-348	49	6	an	an	DET
ijassa-348	49	7	intrusion	intrusion	NOUN
ijassa-348	49	8	detection	detection	NOUN
ijassa-348	49	9	model	model	NOUN
ijassa-348	49	10	by	by	ADP
ijassa-348	49	11	using	use	VERB
ijassa-348	49	12	pca	pca	PROPN
ijassa-348	49	13	as	as	ADP
ijassa-348	49	14	the	the	DET
ijassa-348	49	15	dimensionality	dimensionality	NOUN
ijassa-348	49	16	reduction	reduction	NOUN
ijassa-348	49	17	technique	technique	NOUN
ijassa-348	49	18	and	and	CCONJ
ijassa-348	49	19	svm	svm	NOUN
ijassa-348	49	20	as	as	ADP
ijassa-348	49	21	the	the	DET
ijassa-348	49	22	classifier	classifier	NOUN
ijassa-348	49	23	[	[	X
ijassa-348	49	24	11	11	NUM
ijassa-348	49	25	,	,	PUNCT
ijassa-348	49	26	12	12	NUM
ijassa-348	49	27	]	]	PUNCT
ijassa-348	49	28	.	.	PUNCT
ijassa-348	50	1	the	the	DET
ijassa-348	50	2	kernel	kernel	PROPN
ijassa-348	50	3	parameters	parameter	NOUN
ijassa-348	50	4	of	of	ADP
ijassa-348	50	5	svm	svm	PROPN
ijassa-348	50	6	are	be	AUX
ijassa-348	50	7	optimized	optimize	VERB
ijassa-348	50	8	by	by	ADP
ijassa-348	50	9	considering	consider	VERB
ijassa-348	50	10	the	the	DET
ijassa-348	50	11	variance	variance	NOUN
ijassa-348	50	12	of	of	ADP
ijassa-348	50	13	samples	sample	NOUN
ijassa-348	50	14	available	available	ADJ
ijassa-348	50	15	in	in	ADP
ijassa-348	50	16	the	the	DET
ijassa-348	50	17	same	same	ADJ
ijassa-348	50	18	and	and	CCONJ
ijassa-348	50	19	different	different	ADJ
ijassa-348	50	20	classes	class	NOUN
ijassa-348	50	21	.	.	PUNCT
ijassa-348	51	1	this	this	DET
ijassa-348	51	2	model	model	NOUN
ijassa-348	51	3	provided	provide	VERB
ijassa-348	51	4	a	a	DET
ijassa-348	51	5	better	well	ADJ
ijassa-348	51	6	classification	classification	NOUN
ijassa-348	51	7	accuracy	accuracy	NOUN
ijassa-348	51	8	.	.	PUNCT
ijassa-348	52	1	ajith	ajith	PROPN
ijassa-348	52	2	et	et	PROPN
ijassa-348	52	3	al	al	PROPN
ijassa-348	52	4	.	.	PROPN
ijassa-348	52	5	built	build	VERB
ijassa-348	52	6	a	a	DET
ijassa-348	52	7	light	light	ADJ
ijassa-348	52	8	weight	weight	NOUN
ijassa-348	52	9	ids	id	NOUN
ijassa-348	52	10	using	use	VERB
ijassa-348	52	11	genetic	genetic	ADJ
ijassa-348	52	12	programming	programming	NOUN
ijassa-348	52	13	approaches	approach	NOUN
ijassa-348	52	14	[	[	X
ijassa-348	52	15	13	13	NUM
ijassa-348	52	16	]	]	PUNCT
ijassa-348	52	17	.	.	PUNCT
ijassa-348	53	1	the	the	DET
ijassa-348	53	2	experimental	experimental	ADJ
ijassa-348	53	3	results	result	NOUN
ijassa-348	53	4	proved	prove	VERB
ijassa-348	53	5	that	that	SCONJ
ijassa-348	53	6	the	the	DET
ijassa-348	53	7	accuracy	accuracy	NOUN
ijassa-348	53	8	was	be	AUX
ijassa-348	53	9	better	well	ADJ
ijassa-348	53	10	in	in	ADP
ijassa-348	53	11	comparison	comparison	NOUN
ijassa-348	53	12	to	to	ADP
ijassa-348	53	13	traditional	traditional	ADJ
ijassa-348	53	14	intrusion	intrusion	NOUN
ijassa-348	53	15	detection	detection	NOUN
ijassa-348	53	16	models	model	NOUN
ijassa-348	53	17	.	.	PUNCT
ijassa-348	54	1	kuanga	kuanga	PROPN
ijassa-348	54	2	et	et	PROPN
ijassa-348	54	3	al	al	PROPN
ijassa-348	54	4	.	.	PROPN
ijassa-348	54	5	developed	develop	VERB
ijassa-348	54	6	a	a	DET
ijassa-348	54	7	hybrid	hybrid	ADJ
ijassa-348	54	8	kpca	kpca	NOUN
ijassa-348	54	9	svm	svm	VERB
ijassa-348	54	10	with	with	ADP
ijassa-348	54	11	ga	ga	PROPN
ijassa-348	54	12	model	model	NOUN
ijassa-348	54	13	for	for	ADP
ijassa-348	54	14	intrusion	intrusion	NOUN
ijassa-348	54	15	detection	detection	NOUN
ijassa-348	55	1	[	[	X
ijassa-348	55	2	14	14	NUM
ijassa-348	55	3	]	]	PUNCT
ijassa-348	55	4	.	.	PUNCT
ijassa-348	56	1	the	the	DET
ijassa-348	56	2	authors	author	NOUN
ijassa-348	56	3	used	use	VERB
ijassa-348	56	4	kpca	kpca	NOUN
ijassa-348	56	5	to	to	PART
ijassa-348	56	6	extract	extract	VERB
ijassa-348	56	7	the	the	DET
ijassa-348	56	8	primary	primary	ADJ
ijassa-348	56	9	features	feature	NOUN
ijassa-348	56	10	of	of	ADP
ijassa-348	56	11	intrusion	intrusion	NOUN
ijassa-348	56	12	detection	detection	NOUN
ijassa-348	56	13	dataset	dataset	NOUN
ijassa-348	56	14	.	.	PUNCT
ijassa-348	57	1	svm	svm	PROPN
ijassa-348	57	2	multilayer	multilayer	PROPN
ijassa-348	57	3	model	model	NOUN
ijassa-348	57	4	is	be	AUX
ijassa-348	57	5	employed	employ	VERB
ijassa-348	57	6	as	as	ADP
ijassa-348	57	7	the	the	DET
ijassa-348	57	8	classifier	classifier	NOUN
ijassa-348	57	9	to	to	PART
ijassa-348	57	10	identify	identify	VERB
ijassa-348	57	11	the	the	DET
ijassa-348	57	12	attack	attack	NOUN
ijassa-348	57	13	.	.	PUNCT
ijassa-348	58	1	chebrolu	chebrolu	PROPN
ijassa-348	58	2	et	et	PROPN
ijassa-348	58	3	al	al	PROPN
ijassa-348	58	4	.	.	PROPN
ijassa-348	58	5	evaluated	evaluate	VERB
ijassa-348	58	6	the	the	DET
ijassa-348	58	7	performance	performance	NOUN
ijassa-348	58	8	of	of	ADP
ijassa-348	58	9	two	two	NUM
ijassa-348	58	10	feature	feature	NOUN
ijassa-348	58	11	selection	selection	NOUN
ijassa-348	58	12	techniques	technique	NOUN
ijassa-348	58	13	such	such	ADJ
ijassa-348	58	14	as	as	ADP
ijassa-348	58	15	bayesian	bayesian	NOUN
ijassa-348	58	16	networks	network	NOUN
ijassa-348	58	17	(	(	PUNCT
ijassa-348	58	18	bn	bn	NOUN
ijassa-348	58	19	)	)	PUNCT
ijassa-348	58	20	and	and	CCONJ
ijassa-348	58	21	classification	classification	NOUN
ijassa-348	58	22	and	and	CCONJ
ijassa-348	58	23	regression	regression	NOUN
ijassa-348	58	24	trees	tree	NOUN
ijassa-348	58	25	(	(	PUNCT
ijassa-348	58	26	cart	cart	NOUN
ijassa-348	58	27	)	)	PUNCT
ijassa-348	58	28	and	and	CCONJ
ijassa-348	58	29	also	also	ADV
ijassa-348	58	30	an	an	DET
ijassa-348	58	31	integration	integration	NOUN
ijassa-348	58	32	of	of	ADP
ijassa-348	58	33	cart	cart	NOUN
ijassa-348	58	34	and	and	CCONJ
ijassa-348	58	35	bn	bn	NOUN
ijassa-348	59	1	[	[	X
ijassa-348	59	2	15	15	NUM
ijassa-348	59	3	]	]	PUNCT
ijassa-348	59	4	.	.	PUNCT
ijassa-348	60	1	results	result	NOUN
ijassa-348	60	2	illustrate	illustrate	VERB
ijassa-348	60	3	that	that	SCONJ
ijassa-348	60	4	feature	feature	NOUN
ijassa-348	60	5	selection	selection	NOUN
ijassa-348	60	6	is	be	AUX
ijassa-348	60	7	very	very	ADV
ijassa-348	60	8	effective	effective	ADJ
ijassa-348	60	9	in	in	ADP
ijassa-348	60	10	the	the	DET
ijassa-348	60	11	development	development	NOUN
ijassa-348	60	12	of	of	ADP
ijassa-348	60	13	real	real	ADJ
ijassa-348	60	14	world	world	NOUN
ijassa-348	60	15	intrusion	intrusion	NOUN
ijassa-348	60	16	detection	detection	NOUN
ijassa-348	60	17	models	model	NOUN
ijassa-348	60	18	.	.	PUNCT
ijassa-348	61	1	sandhya	sandhya	PROPN
ijassa-348	61	2	et	et	PROPN
ijassa-348	61	3	al	al	PROPN
ijassa-348	61	4	.	.	PROPN
ijassa-348	61	5	developed	develop	VERB
ijassa-348	61	6	a	a	DET
ijassa-348	61	7	hybrid	hybrid	ADJ
ijassa-348	61	8	intrusion	intrusion	NOUN
ijassa-348	61	9	detection	detection	NOUN
ijassa-348	61	10	model	model	NOUN
ijassa-348	61	11	by	by	ADP
ijassa-348	61	12	combining	combine	VERB
ijassa-348	61	13	decision	decision	NOUN
ijassa-348	61	14	trees	tree	NOUN
ijassa-348	61	15	and	and	CCONJ
ijassa-348	61	16	svm	svm	PROPN
ijassa-348	61	17	and	and	CCONJ
ijassa-348	61	18	also	also	ADV
ijassa-348	61	19	an	an	DET
ijassa-348	61	20	ensemble	ensemble	ADJ
ijassa-348	61	21	of	of	ADP
ijassa-348	61	22	other	other	ADJ
ijassa-348	61	23	base	base	NOUN
ijassa-348	61	24	classifiers	classifier	NOUN
ijassa-348	61	25	[	[	X
ijassa-348	61	26	3	3	NUM
ijassa-348	61	27	]	]	PUNCT
ijassa-348	61	28	.	.	PUNCT
ijassa-348	62	1	this	this	DET
ijassa-348	62	2	hybrid	hybrid	ADJ
ijassa-348	62	3	model	model	NOUN
ijassa-348	62	4	maximized	maximize	VERB
ijassa-348	62	5	accuracy	accuracy	NOUN
ijassa-348	62	6	and	and	CCONJ
ijassa-348	62	7	minimized	minimized	ADJ
ijassa-348	62	8	complexity	complexity	NOUN
ijassa-348	62	9	and	and	CCONJ
ijassa-348	62	10	results	result	NOUN
ijassa-348	62	11	illustrated	illustrate	VERB
ijassa-348	62	12	that	that	SCONJ
ijassa-348	62	13	the	the	DET
ijassa-348	62	14	proposed	propose	VERB
ijassa-348	62	15	model	model	NOUN
ijassa-348	62	16	provided	provide	VERB
ijassa-348	62	17	an	an	DET
ijassa-348	62	18	accurate	accurate	ADJ
ijassa-348	62	19	ids	id	NOUN
ijassa-348	62	20	.	.	PUNCT
ijassa-348	63	1	perin	perin	PROPN
ijassa-348	63	2	et	et	PROPN
ijassa-348	63	3	al	al	PROPN
ijassa-348	63	4	.	.	PROPN
ijassa-348	63	5	built	build	VERB
ijassa-348	63	6	a	a	DET
ijassa-348	63	7	three	three	NUM
ijassa-348	63	8	layer	layer	NOUN
ijassa-348	63	9	multi	multi	NOUN
ijassa-348	63	10	classifier	classifier	PROPN
ijassa-348	63	11	intrusion	intrusion	PROPN
ijassa-348	63	12	detection	detection	NOUN
ijassa-348	63	13	model	model	NOUN
ijassa-348	63	14	to	to	PART
ijassa-348	63	15	increase	increase	VERB
ijassa-348	63	16	the	the	DET
ijassa-348	63	17	overall	overall	ADJ
ijassa-348	63	18	accuracy	accuracy	NOUN
ijassa-348	64	1	[	[	X
ijassa-348	64	2	16	16	NUM
ijassa-348	64	3	]	]	PUNCT
ijassa-348	64	4	.	.	PUNCT
ijassa-348	65	1	the	the	DET
ijassa-348	65	2	performances	performance	NOUN
ijassa-348	65	3	were	be	AUX
ijassa-348	65	4	analyzed	analyze	VERB
ijassa-348	65	5	from	from	ADP
ijassa-348	65	6	a	a	DET
ijassa-348	65	7	vari	vari	PROPN
ijassa-348	65	8	-	-	PUNCT
ijassa-348	65	9	ety	ety	NOUN
ijassa-348	65	10	of	of	ADP
ijassa-348	65	11	combination	combination	NOUN
ijassa-348	65	12	techniques	technique	NOUN
ijassa-348	65	13	such	such	ADJ
ijassa-348	65	14	as	as	ADP
ijassa-348	65	15	fuzzy	fuzzy	ADJ
ijassa-348	65	16	k	k	PROPN
ijassa-348	65	17	-	-	PUNCT
ijassa-348	65	18	nn	nn	PROPN
ijassa-348	65	19	classifier	classifier	NOUN
ijassa-348	65	20	,	,	PUNCT
ijassa-348	65	21	na¨ıve	na¨ıve	PROPN
ijassa-348	65	22	bayes	bayes	PROPN
ijassa-348	65	23	classifier	classifier	NOUN
ijassa-348	65	24	and	and	CCONJ
ijassa-348	65	25	back	back	ADJ
ijassa-348	65	26	propagation	propagation	NOUN
ijassa-348	65	27	neural	neural	ADJ
ijassa-348	65	28	network	network	NOUN
ijassa-348	65	29	classifier	classifier	NOUN
ijassa-348	65	30	and	and	CCONJ
ijassa-348	65	31	the	the	DET
ijassa-348	65	32	decision	decision	NOUN
ijassa-348	65	33	obtained	obtain	VERB
ijassa-348	65	34	from	from	ADP
ijassa-348	65	35	multiple	multiple	ADJ
ijassa-348	65	36	classifiers	classifier	NOUN
ijassa-348	65	37	are	be	AUX
ijassa-348	65	38	combined	combine	VERB
ijassa-348	65	39	into	into	ADP
ijassa-348	65	40	a	a	DET
ijassa-348	65	41	single	single	ADJ
ijassa-348	65	42	result	result	NOUN
ijassa-348	65	43	.	.	PUNCT
ijassa-348	66	1	the	the	DET
ijassa-348	66	2	results	result	NOUN
ijassa-348	66	3	proved	prove	VERB
ijassa-348	66	4	that	that	SCONJ
ijassa-348	66	5	the	the	DET
ijassa-348	66	6	detection	detection	NOUN
ijassa-348	66	7	performance	performance	NOUN
ijassa-348	66	8	is	be	AUX
ijassa-348	66	9	better	well	ADJ
ijassa-348	66	10	than	than	ADP
ijassa-348	66	11	deploying	deploy	VERB
ijassa-348	66	12	a	a	DET
ijassa-348	66	13	single	single	ADJ
ijassa-348	66	14	classifier	classifier	NOUN
ijassa-348	66	15	when	when	SCONJ
ijassa-348	66	16	using	use	VERB
ijassa-348	66	17	a	a	DET
ijassa-348	66	18	full	full	ADJ
ijassa-348	66	19	feature	feature	NOUN
ijassa-348	66	20	set	set	VERB
ijassa-348	66	21	or	or	CCONJ
ijassa-348	66	22	partial	partial	ADJ
ijassa-348	66	23	feature	feature	NOUN
ijassa-348	66	24	set	set	VERB
ijassa-348	66	25	.	.	PUNCT
ijassa-348	67	1	chandra	chandra	NOUN
ijassa-348	67	2	and	and	CCONJ
ijassa-348	67	3	yao	yao	PROPN
ijassa-348	67	4	developed	develop	VERB
ijassa-348	67	5	an	an	DET
ijassa-348	67	6	ensemble	ensemble	ADJ
ijassa-348	67	7	based	base	VERB
ijassa-348	67	8	neural	neural	ADJ
ijassa-348	67	9	network	network	NOUN
ijassa-348	67	10	wherein	wherein	SCONJ
ijassa-348	67	11	the	the	DET
ijassa-348	67	12	outputs	output	NOUN
ijassa-348	67	13	are	be	AUX
ijassa-348	67	14	combined	combine	VERB
ijassa-348	67	15	in	in	ADP
ijassa-348	67	16	a	a	DET
ijassa-348	67	17	form	form	NOUN
ijassa-348	67	18	that	that	PRON
ijassa-348	67	19	resulted	result	VERB
ijassa-348	67	20	in	in	ADP
ijassa-348	67	21	a	a	DET
ijassa-348	67	22	sig	sig	NOUN
ijassa-348	67	23	-	-	PUNCT
ijassa-348	67	24	nificant	nificant	ADJ
ijassa-348	67	25	improvement	improvement	NOUN
ijassa-348	67	26	in	in	ADP
ijassa-348	67	27	the	the	DET
ijassa-348	67	28	generalization	generalization	NOUN
ijassa-348	67	29	performance	performance	NOUN
ijassa-348	67	30	[	[	X
ijassa-348	67	31	17	17	NUM
ijassa-348	67	32	]	]	PUNCT
ijassa-348	67	33	.	.	PUNCT
ijassa-348	68	1	srinivas	srinivas	PROPN
ijassa-348	68	2	et	et	PROPN
ijassa-348	68	3	al	al	PROPN
ijassa-348	68	4	.	.	PROPN
ijassa-348	68	5	built	build	VERB
ijassa-348	68	6	18	18	NUM
ijassa-348	68	7	sumaiya	sumaiya	NOUN
ijassa-348	68	8	thaseen	thaseen	PROPN
ijassa-348	68	9	and	and	CCONJ
ijassa-348	68	10	ch.aswani	ch.aswani	X
ijassa-348	69	1	kumar	kumar	PROPN
ijassa-348	69	2	:	:	PUNCT
ijassa-348	69	3	intrusion	intrusion	NOUN
ijassa-348	69	4	detection	detection	NOUN
ijassa-348	69	5	model	model	NOUN
ijassa-348	69	6	using	use	VERB
ijassa-348	69	7	pca	pca	PROPN
ijassa-348	69	8	and	and	CCONJ
ijassa-348	69	9	...	...	PUNCT
ijassa-348	69	10	an	an	DET
ijassa-348	69	11	ensemble	ensemble	ADJ
ijassa-348	69	12	model	model	NOUN
ijassa-348	69	13	of	of	ADP
ijassa-348	69	14	svm	svm	PROPN
ijassa-348	69	15	,	,	PUNCT
ijassa-348	69	16	artificial	artificial	ADJ
ijassa-348	69	17	neural	neural	ADJ
ijassa-348	69	18	network	network	NOUN
ijassa-348	69	19	(	(	PUNCT
ijassa-348	69	20	ann	ann	PROPN
ijassa-348	69	21	)	)	PUNCT
ijassa-348	69	22	and	and	CCONJ
ijassa-348	69	23	multivariate	multivariate	VERB
ijassa-348	69	24	adaptive	adaptive	ADJ
ijassa-348	69	25	regression	regression	NOUN
ijassa-348	69	26	splines	spline	NOUN
ijassa-348	69	27	(	(	PUNCT
ijassa-348	69	28	mars	mar	NOUN
ijassa-348	69	29	)	)	PUNCT
ijassa-348	69	30	and	and	CCONJ
ijassa-348	69	31	analyzed	analyze	VERB
ijassa-348	69	32	the	the	DET
ijassa-348	69	33	performance	performance	NOUN
ijassa-348	69	34	which	which	PRON
ijassa-348	69	35	was	be	AUX
ijassa-348	69	36	superior	superior	ADJ
ijassa-348	69	37	to	to	ADP
ijassa-348	69	38	individual	individual	ADJ
ijassa-348	69	39	approaches	approach	NOUN
ijassa-348	69	40	with	with	ADP
ijassa-348	69	41	respect	respect	NOUN
ijassa-348	69	42	to	to	ADP
ijassa-348	69	43	classification	classification	NOUN
ijassa-348	69	44	accuracy	accuracy	NOUN
ijassa-348	69	45	[	[	X
ijassa-348	69	46	18	18	NUM
ijassa-348	69	47	]	]	PUNCT
ijassa-348	69	48	.	.	PUNCT
ijassa-348	70	1	syarif	syarif	NOUN
ijassa-348	70	2	et	et	PROPN
ijassa-348	71	1	al	al	PROPN
ijassa-348	71	2	.	.	PROPN
ijassa-348	71	3	improved	improve	VERB
ijassa-348	71	4	the	the	DET
ijassa-348	71	5	accuracy	accuracy	NOUN
ijassa-348	71	6	and	and	CCONJ
ijassa-348	71	7	false	false	ADJ
ijassa-348	71	8	positive	positive	ADJ
ijassa-348	71	9	rate	rate	NOUN
ijassa-348	71	10	of	of	ADP
ijassa-348	71	11	intrusion	intrusion	NOUN
ijassa-348	71	12	detec	detec	ADJ
ijassa-348	71	13	-	-	PUNCT
ijassa-348	71	14	tion	tion	NOUN
ijassa-348	71	15	by	by	ADP
ijassa-348	71	16	constructing	construct	VERB
ijassa-348	71	17	an	an	DET
ijassa-348	71	18	ensemble	ensemble	NOUN
ijassa-348	71	19	of	of	ADP
ijassa-348	71	20	bagging	bagging	NOUN
ijassa-348	71	21	,	,	PUNCT
ijassa-348	71	22	boosting	boost	VERB
ijassa-348	71	23	and	and	CCONJ
ijassa-348	71	24	stacking	stack	VERB
ijassa-348	71	25	[	[	NOUN
ijassa-348	71	26	19	19	NUM
ijassa-348	71	27	]	]	PUNCT
ijassa-348	71	28	.	.	PUNCT
ijassa-348	72	1	the	the	DET
ijassa-348	72	2	base	base	NOUN
ijassa-348	72	3	classifiers	classifier	NOUN
ijassa-348	72	4	for	for	ADP
ijassa-348	72	5	these	these	DET
ijassa-348	72	6	ensemble	ensemble	ADJ
ijassa-348	72	7	models	model	NOUN
ijassa-348	72	8	were	be	AUX
ijassa-348	72	9	na¨ıve	na¨ıve	NUM
ijassa-348	72	10	bayes	baye	NOUN
ijassa-348	72	11	,	,	PUNCT
ijassa-348	72	12	decision	decision	NOUN
ijassa-348	72	13	tree	tree	NOUN
ijassa-348	72	14	,	,	PUNCT
ijassa-348	72	15	rule	rule	NOUN
ijassa-348	72	16	induction	induction	NOUN
ijassa-348	72	17	and	and	CCONJ
ijassa-348	72	18	nearest	near	ADJ
ijassa-348	72	19	neighbor	neighbor	NOUN
ijassa-348	72	20	.	.	PUNCT
ijassa-348	73	1	their	their	PRON
ijassa-348	73	2	results	result	NOUN
ijassa-348	73	3	indicated	indicate	VERB
ijassa-348	73	4	that	that	SCONJ
ijassa-348	73	5	the	the	DET
ijassa-348	73	6	accuracy	accuracy	NOUN
ijassa-348	73	7	for	for	ADP
ijassa-348	73	8	known	know	VERB
ijassa-348	73	9	intrusions	intrusion	NOUN
ijassa-348	73	10	was	be	AUX
ijassa-348	73	11	more	more	ADJ
ijassa-348	73	12	than	than	ADP
ijassa-348	73	13	99	99	NUM
ijassa-348	73	14	%	%	NOUN
ijassa-348	73	15	but	but	CCONJ
ijassa-348	73	16	novel	novel	ADJ
ijassa-348	73	17	intrusions	intrusion	NOUN
ijassa-348	73	18	were	be	AUX
ijassa-348	73	19	identified	identify	VERB
ijassa-348	73	20	with	with	ADP
ijassa-348	73	21	accuracy	accuracy	NOUN
ijassa-348	73	22	levels	level	NOUN
ijassa-348	73	23	of	of	ADP
ijassa-348	73	24	60	60	NUM
ijassa-348	73	25	%	%	NOUN
ijassa-348	73	26	.	.	PUNCT
ijassa-348	74	1	thus	thus	ADV
ijassa-348	74	2	bagging	bag	VERB
ijassa-348	74	3	and	and	CCONJ
ijassa-348	74	4	boosting	boosting	NOUN
ijassa-348	74	5	did	do	AUX
ijassa-348	74	6	not	not	PART
ijassa-348	74	7	improve	improve	VERB
ijassa-348	74	8	the	the	DET
ijassa-348	74	9	accuracy	accuracy	NOUN
ijassa-348	74	10	significantly	significantly	ADV
ijassa-348	74	11	whereas	whereas	SCONJ
ijassa-348	74	12	stacking	stack	VERB
ijassa-348	74	13	decreased	decrease	VERB
ijassa-348	74	14	the	the	DET
ijassa-348	74	15	false	false	ADJ
ijassa-348	74	16	positive	positive	ADJ
ijassa-348	74	17	rate	rate	NOUN
ijassa-348	74	18	by	by	ADP
ijassa-348	74	19	46	46	NUM
ijassa-348	74	20	%	%	NOUN
ijassa-348	74	21	.	.	PUNCT
ijassa-348	75	1	these	these	DET
ijassa-348	75	2	ensembles	ensemble	NOUN
ijassa-348	75	3	increase	increase	VERB
ijassa-348	75	4	the	the	DET
ijassa-348	75	5	execution	execution	NOUN
ijassa-348	75	6	time	time	NOUN
ijassa-348	75	7	and	and	CCONJ
ijassa-348	75	8	hence	hence	ADV
ijassa-348	75	9	are	be	AUX
ijassa-348	75	10	not	not	PART
ijassa-348	75	11	practical	practical	ADJ
ijassa-348	75	12	to	to	PART
ijassa-348	75	13	be	be	AUX
ijassa-348	75	14	implemented	implement	VERB
ijassa-348	75	15	in	in	ADP
ijassa-348	75	16	an	an	DET
ijassa-348	75	17	ids	id	NOUN
ijassa-348	75	18	.	.	PUNCT
ijassa-348	76	1	bahri	bahri	PROPN
ijassa-348	76	2	et	et	PROPN
ijassa-348	76	3	al	al	PROPN
ijassa-348	76	4	built	build	VERB
ijassa-348	76	5	a	a	DET
ijassa-348	76	6	ensemble	ensemble	ADJ
ijassa-348	76	7	method	method	NOUN
ijassa-348	76	8	called	call	VERB
ijassa-348	76	9	greedyboost	greedyboost	PROPN
ijassa-348	76	10	and	and	CCONJ
ijassa-348	76	11	compared	compare	VERB
ijassa-348	76	12	with	with	ADP
ijassa-348	76	13	adaboost	adaboost	ADV
ijassa-348	76	14	and	and	CCONJ
ijassa-348	76	15	c4.5	c4.5	PROPN
ijassa-348	77	1	[	[	X
ijassa-348	77	2	20	20	NUM
ijassa-348	77	3	]	]	PUNCT
ijassa-348	77	4	.	.	PUNCT
ijassa-348	78	1	however	however	ADV
ijassa-348	78	2	the	the	DET
ijassa-348	78	3	base	base	NOUN
ijassa-348	78	4	classifier	classifier	NOUN
ijassa-348	78	5	details	detail	NOUN
ijassa-348	78	6	are	be	AUX
ijassa-348	78	7	not	not	PART
ijassa-348	78	8	specified	specify	VERB
ijassa-348	78	9	in	in	ADP
ijassa-348	78	10	the	the	DET
ijassa-348	78	11	paper	paper	NOUN
ijassa-348	78	12	but	but	CCONJ
ijassa-348	78	13	the	the	DET
ijassa-348	78	14	results	result	NOUN
ijassa-348	78	15	indicated	indicate	VERB
ijassa-348	78	16	that	that	SCONJ
ijassa-348	78	17	greedyboost	greedyboost	VERB
ijassa-348	78	18	scores	score	NOUN
ijassa-348	78	19	a	a	DET
ijassa-348	78	20	higher	high	ADJ
ijassa-348	78	21	precision	precision	NOUN
ijassa-348	78	22	and	and	CCONJ
ijassa-348	78	23	recall	recall	NOUN
ijassa-348	78	24	in	in	ADP
ijassa-348	78	25	comparison	comparison	NOUN
ijassa-348	78	26	to	to	ADP
ijassa-348	78	27	probe	probe	NOUN
ijassa-348	78	28	,	,	PUNCT
ijassa-348	78	29	u2r	u2r	PROPN
ijassa-348	78	30	and	and	CCONJ
ijassa-348	78	31	r2l	r2l	NOUN
ijassa-348	78	32	attacks	attack	NOUN
ijassa-348	78	33	present	present	ADJ
ijassa-348	78	34	in	in	ADP
ijassa-348	78	35	kdd’99	kdd’99	PROPN
ijassa-348	78	36	dataset	dataset	PROPN
ijassa-348	78	37	.	.	PUNCT
ijassa-348	79	1	bukhtoyarov	bukhtoyarov	NOUN
ijassa-348	79	2	et	et	PROPN
ijassa-348	79	3	al	al	PROPN
ijassa-348	79	4	designed	design	VERB
ijassa-348	79	5	neural	neural	ADJ
ijassa-348	79	6	network	network	NOUN
ijassa-348	79	7	classifiers	classifier	NOUN
ijassa-348	79	8	by	by	ADP
ijassa-348	79	9	applying	apply	VERB
ijassa-348	79	10	a	a	DET
ijassa-348	79	11	probabilistic	probabilistic	ADJ
ijassa-348	79	12	approach	approach	NOUN
ijassa-348	79	13	to	to	ADP
ijassa-348	79	14	the	the	DET
ijassa-348	79	15	network	network	NOUN
ijassa-348	79	16	intrusion	intrusion	NOUN
ijassa-348	79	17	detection	detection	NOUN
ijassa-348	79	18	.	.	PUNCT
ijassa-348	80	1	genetic	genetic	ADJ
ijassa-348	80	2	programming	programming	NOUN
ijassa-348	80	3	based	base	VERB
ijassa-348	80	4	ensembling	ensemble	VERB
ijassa-348	80	5	(	(	PUNCT
ijassa-348	80	6	gpen	gpen	NOUN
ijassa-348	80	7	)	)	PUNCT
ijassa-348	80	8	was	be	AUX
ijassa-348	80	9	deployed	deploy	VERB
ijassa-348	80	10	to	to	PART
ijassa-348	80	11	design	design	VERB
ijassa-348	80	12	neural	neural	ADJ
ijassa-348	80	13	network	network	NOUN
ijassa-348	80	14	ensembles	ensemble	NOUN
ijassa-348	81	1	[	[	X
ijassa-348	81	2	21	21	NUM
ijassa-348	81	3	]	]	PUNCT
ijassa-348	81	4	.	.	PUNCT
ijassa-348	82	1	they	they	PRON
ijassa-348	82	2	also	also	ADV
ijassa-348	82	3	analyzed	analyze	VERB
ijassa-348	82	4	with	with	ADP
ijassa-348	82	5	the	the	DET
ijassa-348	82	6	kddcup	kddcup	ADJ
ijassa-348	82	7	1999	1999	NUM
ijassa-348	82	8	dataset	dataset	NOUN
ijassa-348	82	9	and	and	CCONJ
ijassa-348	82	10	classified	classify	VERB
ijassa-348	82	11	the	the	DET
ijassa-348	82	12	attacks	attack	NOUN
ijassa-348	82	13	.	.	PUNCT
ijassa-348	83	1	cordeiro	cordeiro	PROPN
ijassa-348	83	2	and	and	CCONJ
ijassa-348	83	3	pappa	pappa	PROPN
ijassa-348	83	4	utilized	utilize	VERB
ijassa-348	83	5	the	the	DET
ijassa-348	83	6	particle	particle	NOUN
ijassa-348	83	7	swarm	swarm	NOUN
ijassa-348	83	8	optimization	optimization	NOUN
ijassa-348	83	9	(	(	PUNCT
ijassa-348	83	10	pso	pso	NOUN
ijassa-348	83	11	)	)	PUNCT
ijassa-348	83	12	by	by	ADP
ijassa-348	83	13	weighing	weigh	VERB
ijassa-348	83	14	the	the	DET
ijassa-348	83	15	classifications	classification	NOUN
ijassa-348	83	16	obtained	obtain	VERB
ijassa-348	83	17	from	from	ADP
ijassa-348	83	18	different	different	ADJ
ijassa-348	83	19	classifiers	classifier	NOUN
ijassa-348	83	20	[	[	X
ijassa-348	83	21	22	22	NUM
ijassa-348	83	22	]	]	PUNCT
ijassa-348	83	23	.	.	PUNCT
ijassa-348	84	1	the	the	DET
ijassa-348	84	2	four	four	NUM
ijassa-348	84	3	classification	classification	NOUN
ijassa-348	84	4	algorithms	algorithm	NOUN
ijassa-348	84	5	used	use	VERB
ijassa-348	84	6	were	be	AUX
ijassa-348	84	7	knn	knn	PROPN
ijassa-348	84	8	,	,	PUNCT
ijassa-348	84	9	näıve	näıve	PROPN
ijassa-348	84	10	bayes	bayes	PROPN
ijassa-348	84	11	,	,	PUNCT
ijassa-348	84	12	rocchio	rocchio	NOUN
ijassa-348	84	13	and	and	CCONJ
ijassa-348	84	14	svm	svm	ADJ
ijassa-348	84	15	.	.	PUNCT
ijassa-348	85	1	they	they	PRON
ijassa-348	85	2	used	use	VERB
ijassa-348	85	3	datasets	dataset	NOUN
ijassa-348	85	4	of	of	ADP
ijassa-348	85	5	users	user	NOUN
ijassa-348	85	6	of	of	ADP
ijassa-348	85	7	video	video	ADJ
ijassa-348	85	8	social	social	ADJ
ijassa-348	85	9	network	network	NOUN
ijassa-348	85	10	for	for	ADP
ijassa-348	85	11	classification	classification	NOUN
ijassa-348	85	12	.	.	PUNCT
ijassa-348	86	1	their	their	PRON
ijassa-348	86	2	results	result	NOUN
ijassa-348	86	3	outperformed	outperform	VERB
ijassa-348	86	4	the	the	DET
ijassa-348	86	5	single	single	ADJ
ijassa-348	86	6	classifiers	classifier	NOUN
ijassa-348	86	7	.	.	PUNCT
ijassa-348	87	1	the	the	DET
ijassa-348	87	2	motivation	motivation	NOUN
ijassa-348	87	3	for	for	ADP
ijassa-348	87	4	selecting	select	VERB
ijassa-348	87	5	algorithms	algorithm	NOUN
ijassa-348	87	6	in	in	ADP
ijassa-348	87	7	the	the	DET
ijassa-348	87	8	ensemble	ensemble	ADJ
ijassa-348	87	9	is	be	AUX
ijassa-348	87	10	due	due	ADJ
ijassa-348	87	11	to	to	ADP
ijassa-348	87	12	the	the	DET
ijassa-348	87	13	fact	fact	NOUN
ijassa-348	87	14	that	that	SCONJ
ijassa-348	87	15	an	an	DET
ijassa-348	87	16	ensemble	ensemble	ADJ
ijassa-348	87	17	based	base	VERB
ijassa-348	87	18	on	on	ADP
ijassa-348	87	19	the	the	DET
ijassa-348	87	20	four	four	NUM
ijassa-348	87	21	expert	expert	ADJ
ijassa-348	87	22	algorithms	algorithm	NOUN
ijassa-348	87	23	:	:	PUNCT
ijassa-348	87	24	linear	linear	ADJ
ijassa-348	87	25	discriminant	discriminant	NOUN
ijassa-348	87	26	classifier	classifier	NOUN
ijassa-348	87	27	(	(	PUNCT
ijassa-348	87	28	ldc	ldc	PROPN
ijassa-348	87	29	)	)	PUNCT
ijassa-348	87	30	,	,	PUNCT
ijassa-348	87	31	quadratic	quadratic	ADJ
ijassa-348	87	32	discriminant	discriminant	NOUN
ijassa-348	87	33	classifier	classifier	NOUN
ijassa-348	87	34	(	(	PUNCT
ijassa-348	87	35	qdc	qdc	PROPN
ijassa-348	87	36	)	)	PUNCT
ijassa-348	87	37	,	,	PUNCT
ijassa-348	87	38	k	k	X
ijassa-348	87	39	-	-	PUNCT
ijassa-348	87	40	nearest	near	ADJ
ijassa-348	87	41	neighbor	neighbor	NOUN
ijassa-348	87	42	(	(	PUNCT
ijassa-348	87	43	knn	knn	PROPN
ijassa-348	87	44	)	)	PUNCT
ijassa-348	87	45	and	and	CCONJ
ijassa-348	87	46	back	back	ADJ
ijassa-348	87	47	propagation	propagation	NOUN
ijassa-348	87	48	.	.	PUNCT
ijassa-348	88	1	the	the	DET
ijassa-348	88	2	ensemble	ensemble	ADJ
ijassa-348	88	3	is	be	AUX
ijassa-348	88	4	obtained	obtain	VERB
ijassa-348	88	5	by	by	ADP
ijassa-348	88	6	integrating	integrate	VERB
ijassa-348	88	7	the	the	DET
ijassa-348	88	8	experts	expert	NOUN
ijassa-348	88	9	opinion	opinion	NOUN
ijassa-348	88	10	with	with	ADP
ijassa-348	88	11	a	a	DET
ijassa-348	88	12	weight	weight	NOUN
ijassa-348	88	13	coefficient	coefficient	NOUN
ijassa-348	88	14	assigned	assign	VERB
ijassa-348	88	15	by	by	ADP
ijassa-348	88	16	weighted	weight	VERB
ijassa-348	88	17	majority	majority	NOUN
ijassa-348	88	18	voting	voting	NOUN
ijassa-348	88	19	is	be	AUX
ijassa-348	88	20	already	already	ADV
ijassa-348	88	21	tested	test	VERB
ijassa-348	88	22	on	on	ADP
ijassa-348	88	23	four	four	NUM
ijassa-348	88	24	widely	widely	ADV
ijassa-348	88	25	used	use	VERB
ijassa-348	88	26	datasets	dataset	NOUN
ijassa-348	88	27	of	of	ADP
ijassa-348	88	28	hearth	hearth	NOUN
ijassa-348	88	29	,	,	PUNCT
ijassa-348	88	30	diabetes	diabetes	NOUN
ijassa-348	88	31	,	,	PUNCT
ijassa-348	88	32	iris	iris	NOUN
ijassa-348	88	33	and	and	CCONJ
ijassa-348	88	34	transfusion	transfusion	NOUN
ijassa-348	88	35	.	.	PUNCT
ijassa-348	89	1	this	this	DET
ijassa-348	89	2	ensemble	ensemble	ADJ
ijassa-348	89	3	resulted	result	VERB
ijassa-348	89	4	in	in	ADP
ijassa-348	89	5	better	well	ADJ
ijassa-348	89	6	accuracy	accuracy	NOUN
ijassa-348	89	7	in	in	ADP
ijassa-348	89	8	comparison	comparison	NOUN
ijassa-348	89	9	to	to	ADP
ijassa-348	89	10	simple	simple	ADJ
ijassa-348	89	11	majority	majority	NOUN
ijassa-348	89	12	voting	voting	NOUN
ijassa-348	89	13	approach	approach	NOUN
ijassa-348	89	14	,	,	PUNCT
ijassa-348	89	15	mean	mean	INTJ
ijassa-348	89	16	,	,	PUNCT
ijassa-348	89	17	maximum	maximum	ADJ
ijassa-348	89	18	,	,	PUNCT
ijassa-348	89	19	minimum	minimum	ADJ
ijassa-348	89	20	and	and	CCONJ
ijassa-348	89	21	median	median	ADJ
ijassa-348	89	22	combiner	combiner	NOUN
ijassa-348	89	23	[	[	X
ijassa-348	89	24	23	23	NUM
ijassa-348	89	25	]	]	PUNCT
ijassa-348	89	26	.	.	PUNCT
ijassa-348	90	1	thus	thus	ADV
ijassa-348	90	2	many	many	ADJ
ijassa-348	90	3	hybrid	hybrid	ADJ
ijassa-348	90	4	models	model	NOUN
ijassa-348	90	5	using	use	VERB
ijassa-348	90	6	ensemble	ensemble	NOUN
ijassa-348	90	7	of	of	ADP
ijassa-348	90	8	techniques	technique	NOUN
ijassa-348	90	9	for	for	ADP
ijassa-348	90	10	intrusion	intrusion	NOUN
ijassa-348	90	11	detection	detection	NOUN
ijassa-348	90	12	have	have	AUX
ijassa-348	90	13	been	be	AUX
ijassa-348	90	14	developed	develop	VERB
ijassa-348	90	15	but	but	CCONJ
ijassa-348	90	16	the	the	DET
ijassa-348	90	17	major	major	ADJ
ijassa-348	90	18	issue	issue	NOUN
ijassa-348	90	19	in	in	ADP
ijassa-348	90	20	ensemble	ensemble	ADJ
ijassa-348	90	21	approaches	approach	NOUN
ijassa-348	90	22	is	be	AUX
ijassa-348	90	23	the	the	DET
ijassa-348	90	24	models	model	NOUN
ijassa-348	90	25	were	be	AUX
ijassa-348	90	26	built	build	VERB
ijassa-348	90	27	and	and	CCONJ
ijassa-348	90	28	tested	test	VERB
ijassa-348	90	29	only	only	ADV
ijassa-348	90	30	on	on	ADP
ijassa-348	90	31	kdd	kdd	PROPN
ijassa-348	90	32	datasets	dataset	NOUN
ijassa-348	90	33	and	and	CCONJ
ijassa-348	90	34	thus	thus	ADV
ijassa-348	90	35	a	a	DET
ijassa-348	90	36	generic	generic	ADJ
ijassa-348	90	37	model	model	NOUN
ijassa-348	90	38	that	that	PRON
ijassa-348	90	39	can	can	AUX
ijassa-348	90	40	deploy	deploy	VERB
ijassa-348	90	41	and	and	CCONJ
ijassa-348	90	42	test	test	VERB
ijassa-348	90	43	for	for	ADP
ijassa-348	90	44	any	any	DET
ijassa-348	90	45	real	real	ADJ
ijassa-348	90	46	time	time	NOUN
ijassa-348	90	47	datasets	dataset	NOUN
ijassa-348	90	48	is	be	AUX
ijassa-348	90	49	the	the	DET
ijassa-348	90	50	necessity	necessity	NOUN
ijassa-348	90	51	for	for	ADP
ijassa-348	90	52	the	the	DET
ijassa-348	90	53	current	current	ADJ
ijassa-348	90	54	scenario	scenario	NOUN
ijassa-348	90	55	.	.	PUNCT
ijassa-348	91	1	the	the	DET
ijassa-348	91	2	proposed	propose	VERB
ijassa-348	91	3	model	model	NOUN
ijassa-348	91	4	aims	aim	VERB
ijassa-348	91	5	to	to	PART
ijassa-348	91	6	overcome	overcome	VERB
ijassa-348	91	7	the	the	DET
ijassa-348	91	8	issues	issue	NOUN
ijassa-348	91	9	in	in	ADP
ijassa-348	91	10	the	the	DET
ijassa-348	91	11	existing	exist	VERB
ijassa-348	91	12	ensemble	ensemble	ADJ
ijassa-348	91	13	approach	approach	NOUN
ijassa-348	91	14	and	and	CCONJ
ijassa-348	91	15	also	also	ADV
ijassa-348	91	16	with	with	ADP
ijassa-348	91	17	the	the	DET
ijassa-348	91	18	advantage	advantage	NOUN
ijassa-348	91	19	of	of	ADP
ijassa-348	91	20	developing	develop	VERB
ijassa-348	91	21	modular	modular	ADJ
ijassa-348	91	22	structures	structure	NOUN
ijassa-348	91	23	that	that	PRON
ijassa-348	91	24	can	can	AUX
ijassa-348	91	25	have	have	VERB
ijassa-348	91	26	interchangeable	interchangeable	ADJ
ijassa-348	91	27	positions	position	NOUN
ijassa-348	91	28	.	.	PUNCT
ijassa-348	92	1	another	another	DET
ijassa-348	92	2	advantage	advantage	NOUN
ijassa-348	92	3	of	of	ADP
ijassa-348	92	4	our	our	PRON
ijassa-348	92	5	proposed	propose	VERB
ijassa-348	92	6	ensemble	ensemble	ADJ
ijassa-348	92	7	deadvances	deadvance	NOUN
ijassa-348	92	8	in	in	ADP
ijassa-348	92	9	systems	system	NOUN
ijassa-348	92	10	science	science	NOUN
ijassa-348	92	11	and	and	CCONJ
ijassa-348	92	12	application	application	NOUN
ijassa-348	92	13	(	(	PUNCT
ijassa-348	92	14	2016	2016	NUM
ijassa-348	92	15	)	)	PUNCT
ijassa-348	92	16	vol.16	vol.16	PROPN
ijassa-348	92	17	no.2	no.2	PROPN
ijassa-348	92	18	19	19	NUM
ijassa-348	92	19	signs	sign	NOUN
ijassa-348	92	20	is	be	AUX
ijassa-348	92	21	that	that	SCONJ
ijassa-348	92	22	the	the	DET
ijassa-348	92	23	algorithms	algorithm	NOUN
ijassa-348	92	24	can	can	AUX
ijassa-348	92	25	be	be	AUX
ijassa-348	92	26	replaced	replace	VERB
ijassa-348	92	27	anytime	anytime	ADV
ijassa-348	92	28	with	with	ADP
ijassa-348	92	29	a	a	DET
ijassa-348	92	30	more	more	ADV
ijassa-348	92	31	precise	precise	ADJ
ijassa-348	92	32	one	one	NUM
ijassa-348	92	33	.	.	PUNCT
ijassa-348	93	1	hence	hence	ADV
ijassa-348	93	2	the	the	DET
ijassa-348	93	3	approach	approach	NOUN
ijassa-348	93	4	aims	aim	VERB
ijassa-348	93	5	to	to	PART
ijassa-348	93	6	improve	improve	VERB
ijassa-348	93	7	intrusion	intrusion	NOUN
ijassa-348	93	8	detection	detection	NOUN
ijassa-348	93	9	accuracy	accuracy	NOUN
ijassa-348	93	10	using	use	VERB
ijassa-348	93	11	simple	simple	ADJ
ijassa-348	93	12	techniques	technique	NOUN
ijassa-348	93	13	in	in	ADP
ijassa-348	93	14	ensemble	ensemble	ADJ
ijassa-348	93	15	learning	learning	NOUN
ijassa-348	93	16	integrated	integrate	VERB
ijassa-348	93	17	with	with	ADP
ijassa-348	93	18	pca	pca	PROPN
ijassa-348	93	19	as	as	ADP
ijassa-348	93	20	a	a	DET
ijassa-348	93	21	dimensionality	dimensionality	NOUN
ijassa-348	93	22	reduction	reduction	NOUN
ijassa-348	93	23	technique	technique	NOUN
ijassa-348	93	24	.	.	PUNCT
ijassa-348	94	1	3	3	NUM
ijassa-348	94	2	background	background	NOUN
ijassa-348	94	3	3.1	3.1	NUM
ijassa-348	94	4	preprocessing	preprocessing	NOUN
ijassa-348	94	5	data	datum	NOUN
ijassa-348	94	6	preprocessing	preprocessing	NOUN
ijassa-348	94	7	is	be	AUX
ijassa-348	94	8	very	very	ADV
ijassa-348	94	9	essential	essential	ADJ
ijassa-348	94	10	for	for	ADP
ijassa-348	94	11	huge	huge	ADJ
ijassa-348	94	12	data	datum	NOUN
ijassa-348	94	13	such	such	ADJ
ijassa-348	94	14	as	as	ADP
ijassa-348	94	15	network	network	NOUN
ijassa-348	94	16	traffic	traffic	NOUN
ijassa-348	94	17	.	.	PUNCT
ijassa-348	95	1	reduction	reduction	NOUN
ijassa-348	95	2	of	of	ADP
ijassa-348	95	3	redundant	redundant	ADJ
ijassa-348	95	4	data	datum	NOUN
ijassa-348	95	5	and	and	CCONJ
ijassa-348	95	6	normalization	normalization	NOUN
ijassa-348	95	7	are	be	AUX
ijassa-348	95	8	essential	essential	ADJ
ijassa-348	95	9	to	to	PART
ijassa-348	95	10	be	be	AUX
ijassa-348	95	11	performed	perform	VERB
ijassa-348	95	12	in	in	ADP
ijassa-348	95	13	preprocessing	preprocesse	VERB
ijassa-348	95	14	to	to	PART
ijassa-348	95	15	build	build	VERB
ijassa-348	95	16	a	a	DET
ijassa-348	95	17	balanced	balanced	ADJ
ijassa-348	95	18	set	set	NOUN
ijassa-348	95	19	of	of	ADP
ijassa-348	95	20	data	data	PROPN
ijassa-348	95	21	.	.	PUNCT
ijassa-348	96	1	normalization	normalization	NOUN
ijassa-348	96	2	is	be	AUX
ijassa-348	96	3	the	the	DET
ijassa-348	96	4	process	process	NOUN
ijassa-348	96	5	of	of	ADP
ijassa-348	96	6	transforming	transform	VERB
ijassa-348	96	7	the	the	DET
ijassa-348	96	8	data	datum	NOUN
ijassa-348	96	9	within	within	ADP
ijassa-348	96	10	a	a	DET
ijassa-348	96	11	small	small	ADJ
ijassa-348	96	12	specified	specified	ADJ
ijassa-348	96	13	range	range	NOUN
ijassa-348	96	14	.	.	PUNCT
ijassa-348	97	1	the	the	DET
ijassa-348	97	2	different	different	ADJ
ijassa-348	97	3	normalization	normalization	NOUN
ijassa-348	97	4	techniques	technique	NOUN
ijassa-348	97	5	are	be	AUX
ijassa-348	97	6	min	min	ADJ
ijassa-348	97	7	-	-	ADJ
ijassa-348	97	8	max	max	NOUN
ijassa-348	97	9	normalization	normalization	NOUN
ijassa-348	97	10	,	,	PUNCT
ijassa-348	97	11	z	z	NOUN
ijassa-348	97	12	-	-	PUNCT
ijassa-348	97	13	score	score	NOUN
ijassa-348	97	14	normalization	normalization	NOUN
ijassa-348	97	15	and	and	CCONJ
ijassa-348	97	16	normalization	normalization	NOUN
ijassa-348	97	17	by	by	ADP
ijassa-348	97	18	decimal	decimal	ADJ
ijassa-348	97	19	scaling	scaling	NOUN
ijassa-348	97	20	.	.	PUNCT
ijassa-348	98	1	we	we	PRON
ijassa-348	98	2	select	select	VERB
ijassa-348	98	3	z	z	ADJ
ijassa-348	98	4	-	-	PUNCT
ijassa-348	98	5	score	score	NOUN
ijassa-348	98	6	technique	technique	NOUN
ijassa-348	98	7	because	because	SCONJ
ijassa-348	98	8	it	it	PRON
ijassa-348	98	9	considers	consider	VERB
ijassa-348	98	10	the	the	DET
ijassa-348	98	11	mean	mean	ADJ
ijassa-348	98	12	and	and	CCONJ
ijassa-348	98	13	standard	standard	ADJ
ijassa-348	98	14	deviation	deviation	NOUN
ijassa-348	98	15	of	of	ADP
ijassa-348	98	16	the	the	DET
ijassa-348	98	17	attribute	attribute	NOUN
ijassa-348	98	18	.	.	PUNCT
ijassa-348	99	1	d1	d1	PROPN
ijassa-348	99	2	=	=	SYM
ijassa-348	99	3	b−mean	b−mean	X
ijassa-348	99	4	(	(	PUNCT
ijassa-348	99	5	f	f	X
ijassa-348	99	6	)	)	PUNCT
ijassa-348	99	7	std	std	NOUN
ijassa-348	99	8	(	(	PUNCT
ijassa-348	99	9	f	f	X
ijassa-348	99	10	)	)	PUNCT
ijassa-348	99	11	(	(	PUNCT
ijassa-348	99	12	1	1	X
ijassa-348	99	13	)	)	PUNCT
ijassa-348	99	14	where	where	SCONJ
ijassa-348	99	15	,	,	PUNCT
ijassa-348	99	16	mean(f)=	mean(f)=	PROPN
ijassa-348	99	17	sum	sum	NOUN
ijassa-348	99	18	of	of	ADP
ijassa-348	99	19	all	all	DET
ijassa-348	99	20	attribute	attribute	NOUN
ijassa-348	99	21	values	value	NOUN
ijassa-348	99	22	of	of	ADP
ijassa-348	99	23	f	f	PROPN
ijassa-348	99	24	std(f	std(f	PROPN
ijassa-348	99	25	)	)	PUNCT
ijassa-348	99	26	=	=	NOUN
ijassa-348	99	27	standard	standard	ADJ
ijassa-348	99	28	deviation	deviation	NOUN
ijassa-348	99	29	of	of	ADP
ijassa-348	99	30	all	all	DET
ijassa-348	99	31	values	value	NOUN
ijassa-348	99	32	of	of	ADP
ijassa-348	99	33	f.	f.	PROPN
ijassa-348	99	34	3.2	3.2	NUM
ijassa-348	99	35	dimensionality	dimensionality	NOUN
ijassa-348	99	36	reduction	reduction	NOUN
ijassa-348	99	37	dimensionality	dimensionality	NOUN
ijassa-348	99	38	reduction	reduction	NOUN
ijassa-348	99	39	transforms	transform	VERB
ijassa-348	99	40	the	the	DET
ijassa-348	99	41	data	datum	NOUN
ijassa-348	99	42	in	in	ADP
ijassa-348	99	43	the	the	DET
ijassa-348	99	44	high	high	ADJ
ijassa-348	99	45	dimensional	dimensional	ADJ
ijassa-348	99	46	space	space	NOUN
ijassa-348	99	47	to	to	ADP
ijassa-348	99	48	a	a	DET
ijassa-348	99	49	lower	low	ADJ
ijassa-348	99	50	dimension	dimension	NOUN
ijassa-348	99	51	.	.	PUNCT
ijassa-348	100	1	pca	pca	NOUN
ijassa-348	100	2	performs	perform	VERB
ijassa-348	100	3	a	a	DET
ijassa-348	100	4	linear	linear	ADJ
ijassa-348	100	5	mapping	mapping	NOUN
ijassa-348	100	6	of	of	ADP
ijassa-348	100	7	the	the	DET
ijassa-348	100	8	data	datum	NOUN
ijassa-348	100	9	to	to	ADP
ijassa-348	100	10	a	a	DET
ijassa-348	100	11	lower	low	ADJ
ijassa-348	100	12	dimension	dimension	NOUN
ijassa-348	100	13	such	such	ADJ
ijassa-348	100	14	that	that	SCONJ
ijassa-348	100	15	the	the	DET
ijassa-348	100	16	maximum	maximum	ADJ
ijassa-348	100	17	variance	variance	NOUN
ijassa-348	100	18	for	for	ADP
ijassa-348	100	19	the	the	DET
ijassa-348	100	20	data	datum	NOUN
ijassa-348	100	21	is	be	AUX
ijassa-348	100	22	obtained	obtain	VERB
ijassa-348	100	23	.	.	PUNCT
ijassa-348	101	1	the	the	DET
ijassa-348	101	2	procedure	procedure	NOUN
ijassa-348	101	3	begins	begin	VERB
ijassa-348	101	4	with	with	ADP
ijassa-348	101	5	the	the	DET
ijassa-348	101	6	construction	construction	NOUN
ijassa-348	101	7	of	of	ADP
ijassa-348	101	8	correlation	correlation	NOUN
ijassa-348	101	9	matrix	matrix	NOUN
ijassa-348	101	10	and	and	CCONJ
ijassa-348	101	11	computation	computation	NOUN
ijassa-348	101	12	of	of	ADP
ijassa-348	101	13	eigen	eigen	PROPN
ijassa-348	101	14	vectors	vector	NOUN
ijassa-348	101	15	.	.	PUNCT
ijassa-348	102	1	the	the	DET
ijassa-348	102	2	eigen	eigen	PROPN
ijassa-348	102	3	vectors	vector	NOUN
ijassa-348	102	4	that	that	PRON
ijassa-348	102	5	correspond	correspond	VERB
ijassa-348	102	6	to	to	ADP
ijassa-348	102	7	the	the	DET
ijassa-348	102	8	highest	high	ADJ
ijassa-348	102	9	eigen	eigen	PROPN
ijassa-348	102	10	values	value	NOUN
ijassa-348	102	11	will	will	AUX
ijassa-348	102	12	be	be	AUX
ijassa-348	102	13	deployed	deploy	VERB
ijassa-348	102	14	to	to	PART
ijassa-348	102	15	reconstruct	reconstruct	VERB
ijassa-348	102	16	the	the	DET
ijassa-348	102	17	variance	variance	NOUN
ijassa-348	102	18	of	of	ADP
ijassa-348	102	19	the	the	DET
ijassa-348	102	20	original	original	ADJ
ijassa-348	102	21	data	datum	NOUN
ijassa-348	102	22	.	.	PUNCT
ijassa-348	103	1	transformation	transformation	NOUN
ijassa-348	103	2	thus	thus	ADV
ijassa-348	103	3	the	the	DET
ijassa-348	103	4	original	original	ADJ
ijassa-348	103	5	space	space	NOUN
ijassa-348	103	6	is	be	AUX
ijassa-348	103	7	reduced	reduce	VERB
ijassa-348	103	8	to	to	ADP
ijassa-348	103	9	the	the	DET
ijassa-348	103	10	space	space	NOUN
ijassa-348	103	11	obtained	obtain	VERB
ijassa-348	103	12	by	by	ADP
ijassa-348	103	13	a	a	DET
ijassa-348	103	14	few	few	ADJ
ijassa-348	103	15	eigen	eigen	PROPN
ijassa-348	103	16	vectors	vector	NOUN
ijassa-348	103	17	.	.	PUNCT
ijassa-348	104	1	the	the	DET
ijassa-348	104	2	advantages	advantage	NOUN
ijassa-348	104	3	of	of	ADP
ijassa-348	104	4	using	use	VERB
ijassa-348	104	5	dimensionality	dimensionality	NOUN
ijassa-348	104	6	reduction	reduction	NOUN
ijassa-348	104	7	are	be	AUX
ijassa-348	104	8	as	as	SCONJ
ijassa-348	104	9	follows	follow	VERB
ijassa-348	104	10	:	:	PUNCT
ijassa-348	104	11	•	•	NUM
ijassa-348	104	12	time	time	NOUN
ijassa-348	104	13	and	and	CCONJ
ijassa-348	104	14	storage	storage	NOUN
ijassa-348	104	15	space	space	NOUN
ijassa-348	104	16	is	be	AUX
ijassa-348	104	17	reduced	reduce	VERB
ijassa-348	104	18	.	.	PUNCT
ijassa-348	105	1	•	•	NUM
ijassa-348	105	2	improves	improve	VERB
ijassa-348	105	3	the	the	DET
ijassa-348	105	4	performance	performance	NOUN
ijassa-348	105	5	of	of	ADP
ijassa-348	105	6	the	the	DET
ijassa-348	105	7	machine	machine	NOUN
ijassa-348	105	8	learning	learn	VERB
ijassa-348	105	9	model	model	NOUN
ijassa-348	105	10	.	.	PUNCT
ijassa-348	106	1	•	•	NUM
ijassa-348	106	2	visualization	visualization	NOUN
ijassa-348	106	3	of	of	ADP
ijassa-348	106	4	data	datum	NOUN
ijassa-348	106	5	is	be	AUX
ijassa-348	106	6	easier	easy	ADJ
ijassa-348	106	7	as	as	SCONJ
ijassa-348	106	8	the	the	DET
ijassa-348	106	9	dimensions	dimension	NOUN
ijassa-348	106	10	are	be	AUX
ijassa-348	106	11	reduced	reduce	VERB
ijassa-348	106	12	to	to	ADP
ijassa-348	106	13	2d	2d	NUM
ijassa-348	106	14	or	or	CCONJ
ijassa-348	106	15	3d	3d	NUM
ijassa-348	106	16	.	.	PUNCT
ijassa-348	107	1	3.3	3.3	NUM
ijassa-348	107	2	datasets	dataset	NOUN
ijassa-348	107	3	we	we	PRON
ijassa-348	107	4	have	have	AUX
ijassa-348	107	5	analyzed	analyze	VERB
ijassa-348	107	6	the	the	DET
ijassa-348	107	7	knowledge	knowledge	NOUN
ijassa-348	107	8	discovery	discovery	NOUN
ijassa-348	107	9	and	and	CCONJ
ijassa-348	107	10	data	datum	NOUN
ijassa-348	107	11	mining	mining	NOUN
ijassa-348	107	12	1999	1999	NUM
ijassa-348	107	13	standard	standard	NOUN
ijassa-348	107	14	dataset	dataset	NOUN
ijassa-348	107	15	such	such	ADJ
ijassa-348	107	16	as	as	ADP
ijassa-348	107	17	nsl	nsl	NOUN
ijassa-348	107	18	-	-	PUNCT
ijassa-348	107	19	kdd	kdd	PROPN
ijassa-348	107	20	data	datum	NOUN
ijassa-348	107	21	set	set	NOUN
ijassa-348	107	22	which	which	PRON
ijassa-348	107	23	is	be	AUX
ijassa-348	107	24	widely	widely	ADV
ijassa-348	107	25	used	use	VERB
ijassa-348	107	26	as	as	ADP
ijassa-348	107	27	intrusion	intrusion	NOUN
ijassa-348	107	28	detection	detection	NOUN
ijassa-348	107	29	benchmark	benchmark	NOUN
ijassa-348	107	30	datasets	dataset	NOUN
ijassa-348	107	31	.	.	PUNCT
ijassa-348	108	1	the	the	DET
ijassa-348	108	2	nsl	nsl	PROPN
ijassa-348	108	3	-	-	PUNCT
ijassa-348	108	4	kdd	kdd	PROPN
ijassa-348	108	5	data	datum	NOUN
ijassa-348	108	6	set	set	NOUN
ijassa-348	108	7	contains	contain	VERB
ijassa-348	108	8	roughly	roughly	ADV
ijassa-348	108	9	33,300	33,300	NUM
ijassa-348	108	10	samples	sample	NOUN
ijassa-348	108	11	.	.	PUNCT
ijassa-348	109	1	this	this	DET
ijassa-348	109	2	dataset	dataset	NOUN
ijassa-348	109	3	is	be	AUX
ijassa-348	109	4	chosen	choose	VERB
ijassa-348	109	5	because	because	SCONJ
ijassa-348	109	6	of	of	ADP
ijassa-348	109	7	the	the	DET
ijassa-348	109	8	following	follow	VERB
ijassa-348	109	9	benefits	benefit	NOUN
ijassa-348	109	10	[	[	X
ijassa-348	109	11	24	24	NUM
ijassa-348	109	12	]	]	SYM
ijassa-348	109	13	:	:	PUNCT
ijassa-348	109	14	1	1	X
ijassa-348	109	15	)	)	PUNCT
ijassa-348	109	16	no	no	DET
ijassa-348	109	17	redundant	redundant	ADJ
ijassa-348	109	18	records	record	NOUN
ijassa-348	109	19	in	in	ADP
ijassa-348	109	20	the	the	DET
ijassa-348	109	21	training	training	NOUN
ijassa-348	109	22	set	set	NOUN
ijassa-348	109	23	.	.	PUNCT
ijassa-348	110	1	2	2	NUM
ijassa-348	110	2	)	)	PUNCT
ijassa-348	110	3	due	due	ADP
ijassa-348	110	4	to	to	ADP
ijassa-348	110	5	the	the	DET
ijassa-348	110	6	reduction	reduction	NOUN
ijassa-348	110	7	in	in	ADP
ijassa-348	110	8	the	the	DET
ijassa-348	110	9	number	number	NOUN
ijassa-348	110	10	of	of	ADP
ijassa-348	110	11	records	record	NOUN
ijassa-348	110	12	,	,	PUNCT
ijassa-348	110	13	the	the	DET
ijassa-348	110	14	complete	complete	ADJ
ijassa-348	110	15	data	datum	NOUN
ijassa-348	110	16	20	20	NUM
ijassa-348	110	17	sumaiya	sumaiya	NOUN
ijassa-348	110	18	thaseen	thaseen	NOUN
ijassa-348	110	19	and	and	CCONJ
ijassa-348	110	20	ch.aswani	ch.aswani	X
ijassa-348	111	1	kumar	kumar	PROPN
ijassa-348	111	2	:	:	PUNCT
ijassa-348	111	3	intrusion	intrusion	NOUN
ijassa-348	111	4	detection	detection	NOUN
ijassa-348	111	5	model	model	NOUN
ijassa-348	111	6	using	use	VERB
ijassa-348	111	7	pca	pca	PROPN
ijassa-348	111	8	and	and	CCONJ
ijassa-348	111	9	...	...	PUNCT
ijassa-348	111	10	can	can	AUX
ijassa-348	111	11	be	be	AUX
ijassa-348	111	12	used	use	VERB
ijassa-348	111	13	for	for	ADP
ijassa-348	111	14	both	both	DET
ijassa-348	111	15	training	training	NOUN
ijassa-348	111	16	and	and	CCONJ
ijassa-348	111	17	testing	testing	NOUN
ijassa-348	111	18	.	.	PUNCT
ijassa-348	112	1	each	each	DET
ijassa-348	112	2	packet	packet	NOUN
ijassa-348	112	3	in	in	ADP
ijassa-348	112	4	the	the	DET
ijassa-348	112	5	dataset	dataset	NOUN
ijassa-348	112	6	can	can	AUX
ijassa-348	112	7	be	be	AUX
ijassa-348	112	8	classified	classify	VERB
ijassa-348	112	9	in	in	ADP
ijassa-348	112	10	any	any	DET
ijassa-348	112	11	one	one	NUM
ijassa-348	112	12	of	of	ADP
ijassa-348	112	13	the	the	DET
ijassa-348	112	14	classes	class	NOUN
ijassa-348	112	15	namely	namely	ADV
ijassa-348	112	16	normal	normal	ADJ
ijassa-348	112	17	,	,	PUNCT
ijassa-348	112	18	dos	do	NOUN
ijassa-348	112	19	,	,	PUNCT
ijassa-348	112	20	u2r	u2r	NOUN
ijassa-348	112	21	,	,	PUNCT
ijassa-348	112	22	r2l	r2l	NOUN
ijassa-348	112	23	and	and	CCONJ
ijassa-348	112	24	probe	probe	NOUN
ijassa-348	112	25	.	.	PUNCT
ijassa-348	113	1	all	all	DET
ijassa-348	113	2	the	the	DET
ijassa-348	113	3	class	class	NOUN
ijassa-348	113	4	labels	label	NOUN
ijassa-348	113	5	except	except	SCONJ
ijassa-348	113	6	normal	normal	ADJ
ijassa-348	113	7	indicate	indicate	VERB
ijassa-348	113	8	the	the	DET
ijassa-348	113	9	different	different	ADJ
ijassa-348	113	10	attacks	attack	NOUN
ijassa-348	113	11	in	in	ADP
ijassa-348	113	12	the	the	DET
ijassa-348	113	13	dataset	dataset	NOUN
ijassa-348	113	14	.	.	PUNCT
ijassa-348	114	1	the	the	DET
ijassa-348	114	2	other	other	ADJ
ijassa-348	114	3	dataset	dataset	NOUN
ijassa-348	114	4	analyzed	analyze	VERB
ijassa-348	114	5	in	in	ADP
ijassa-348	114	6	our	our	PRON
ijassa-348	114	7	model	model	NOUN
ijassa-348	114	8	is	be	AUX
ijassa-348	114	9	unsw	unsw	PROPN
ijassa-348	114	10	-	-	PUNCT
ijassa-348	114	11	nb	nb	NOUN
ijassa-348	114	12	dataset	dataset	PROPN
ijassa-348	114	13	obtained	obtain	VERB
ijassa-348	114	14	from	from	ADP
ijassa-348	114	15	the	the	DET
ijassa-348	114	16	cyber	cyber	NOUN
ijassa-348	114	17	range	range	NOUN
ijassa-348	114	18	lab	lab	NOUN
ijassa-348	114	19	of	of	ADP
ijassa-348	114	20	the	the	DET
ijassa-348	114	21	australian	australian	ADJ
ijassa-348	114	22	center	center	NOUN
ijassa-348	114	23	for	for	ADP
ijassa-348	114	24	cyber	cyber	ADJ
ijassa-348	114	25	security	security	NOUN
ijassa-348	114	26	(	(	PUNCT
ijassa-348	114	27	accs	accs	PROPN
ijassa-348	114	28	)	)	PUNCT
ijassa-348	114	29	.	.	PUNCT
ijassa-348	115	1	this	this	DET
ijassa-348	115	2	dataset	dataset	NOUN
ijassa-348	115	3	contains	contain	VERB
ijassa-348	115	4	nearly	nearly	ADV
ijassa-348	115	5	1,56,000	1,56,000	NUM
ijassa-348	115	6	samples	sample	NOUN
ijassa-348	115	7	falling	fall	VERB
ijassa-348	115	8	in	in	ADP
ijassa-348	115	9	one	one	NUM
ijassa-348	115	10	of	of	ADP
ijassa-348	115	11	the	the	DET
ijassa-348	115	12	nine	nine	NUM
ijassa-348	115	13	classifica	classifica	PROPN
ijassa-348	115	14	-	-	PUNCT
ijassa-348	115	15	tion	tion	NOUN
ijassa-348	115	16	categories	category	NOUN
ijassa-348	115	17	namely	namely	ADV
ijassa-348	115	18	normal	normal	ADJ
ijassa-348	115	19	,	,	PUNCT
ijassa-348	115	20	analysis	analysis	NOUN
ijassa-348	115	21	,	,	PUNCT
ijassa-348	115	22	backdoor	backdoor	NOUN
ijassa-348	115	23	,	,	PUNCT
ijassa-348	115	24	reconnaissance	reconnaissance	NOUN
ijassa-348	115	25	,	,	PUNCT
ijassa-348	115	26	exploits	exploit	NOUN
ijassa-348	115	27	,	,	PUNCT
ijassa-348	115	28	fuzzers	fuzzer	NOUN
ijassa-348	115	29	,	,	PUNCT
ijassa-348	115	30	generic	generic	ADJ
ijassa-348	115	31	,	,	PUNCT
ijassa-348	115	32	dos	do	NOUN
ijassa-348	115	33	and	and	CCONJ
ijassa-348	115	34	shellcode	shellcode	NOUN
ijassa-348	115	35	.	.	PUNCT
ijassa-348	116	1	this	this	DET
ijassa-348	116	2	dataset	dataset	NOUN
ijassa-348	116	3	has	have	VERB
ijassa-348	116	4	the	the	DET
ijassa-348	116	5	advantages	advantage	NOUN
ijassa-348	116	6	of	of	ADP
ijassa-348	116	7	con	con	NOUN
ijassa-348	116	8	-	-	PUNCT
ijassa-348	116	9	taining	taine	VERB
ijassa-348	116	10	current	current	ADJ
ijassa-348	116	11	attacks	attack	NOUN
ijassa-348	116	12	in	in	ADP
ijassa-348	116	13	the	the	DET
ijassa-348	116	14	network	network	NOUN
ijassa-348	116	15	domain	domain	NOUN
ijassa-348	116	16	.	.	PUNCT
ijassa-348	117	1	3.4	3.4	NUM
ijassa-348	117	2	support	support	NOUN
ijassa-348	117	3	vector	vector	NOUN
ijassa-348	117	4	machine	machine	NOUN
ijassa-348	117	5	support	support	NOUN
ijassa-348	117	6	vector	vector	NOUN
ijassa-348	117	7	machines	machine	NOUN
ijassa-348	117	8	are	be	AUX
ijassa-348	117	9	widely	widely	ADV
ijassa-348	117	10	used	use	VERB
ijassa-348	117	11	for	for	ADP
ijassa-348	117	12	classification	classification	NOUN
ijassa-348	117	13	and	and	CCONJ
ijassa-348	117	14	regression	regression	NOUN
ijassa-348	117	15	problems	problem	NOUN
ijassa-348	117	16	.	.	PUNCT
ijassa-348	118	1	svm	svm	PROPN
ijassa-348	118	2	is	be	AUX
ijassa-348	118	3	preferred	prefer	VERB
ijassa-348	118	4	over	over	ADP
ijassa-348	118	5	other	other	ADJ
ijassa-348	118	6	techniques	technique	NOUN
ijassa-348	118	7	due	due	ADP
ijassa-348	118	8	to	to	ADP
ijassa-348	118	9	the	the	DET
ijassa-348	118	10	low	low	ADJ
ijassa-348	118	11	generalization	generalization	NOUN
ijassa-348	118	12	error	error	NOUN
ijassa-348	118	13	and	and	CCONJ
ijassa-348	118	14	less	less	ADJ
ijassa-348	118	15	over	over	ADP
ijassa-348	118	16	fitting	fitting	ADJ
ijassa-348	118	17	issues	issue	NOUN
ijassa-348	118	18	that	that	PRON
ijassa-348	118	19	arise	arise	VERB
ijassa-348	118	20	from	from	ADP
ijassa-348	118	21	the	the	DET
ijassa-348	118	22	training	training	NOUN
ijassa-348	118	23	data	datum	NOUN
ijassa-348	118	24	set	set	VERB
ijassa-348	118	25	.	.	PUNCT
ijassa-348	119	1	there	there	PRON
ijassa-348	119	2	exist	exist	VERB
ijassa-348	119	3	a	a	DET
ijassa-348	119	4	possibility	possibility	NOUN
ijassa-348	119	5	of	of	ADP
ijassa-348	119	6	high	high	ADJ
ijassa-348	119	7	generalization	generalization	NOUN
ijassa-348	119	8	error	error	NOUN
ijassa-348	119	9	or	or	CCONJ
ijassa-348	119	10	overfitting	overfitte	VERB
ijassa-348	119	11	if	if	SCONJ
ijassa-348	119	12	the	the	DET
ijassa-348	119	13	model	model	NOUN
ijassa-348	119	14	does	do	AUX
ijassa-348	119	15	nt	not	PART
ijassa-348	119	16	scale	scale	VERB
ijassa-348	119	17	well	well	ADV
ijassa-348	119	18	on	on	ADP
ijassa-348	119	19	instances	instance	NOUN
ijassa-348	119	20	not	not	PART
ijassa-348	119	21	available	available	ADJ
ijassa-348	119	22	in	in	ADP
ijassa-348	119	23	the	the	DET
ijassa-348	119	24	training	training	NOUN
ijassa-348	119	25	set	set	NOUN
ijassa-348	119	26	.	.	PUNCT
ijassa-348	120	1	svm	svm	PROPN
ijassa-348	120	2	is	be	AUX
ijassa-348	120	3	very	very	ADV
ijassa-348	120	4	effective	effective	ADJ
ijassa-348	120	5	on	on	ADP
ijassa-348	120	6	data	data	NOUN
ijassa-348	120	7	samples	sample	NOUN
ijassa-348	120	8	that	that	PRON
ijassa-348	120	9	are	be	AUX
ijassa-348	120	10	separable	separable	ADJ
ijassa-348	120	11	in	in	ADP
ijassa-348	120	12	a	a	DET
ijassa-348	120	13	linear	linear	ADJ
ijassa-348	120	14	fashion	fashion	NOUN
ijassa-348	120	15	.	.	PUNCT
ijassa-348	121	1	the	the	DET
ijassa-348	121	2	objective	objective	NOUN
ijassa-348	121	3	is	be	AUX
ijassa-348	121	4	to	to	PART
ijassa-348	121	5	identify	identify	VERB
ijassa-348	121	6	the	the	DET
ijassa-348	121	7	hyperplane	hyperplane	PROPN
ijassa-348	121	8	h	h	NOUN
ijassa-348	121	9	that	that	PRON
ijassa-348	121	10	can	can	AUX
ijassa-348	121	11	split	split	VERB
ijassa-348	121	12	the	the	DET
ijassa-348	121	13	instances	instance	NOUN
ijassa-348	121	14	into	into	ADP
ijassa-348	121	15	two	two	NUM
ijassa-348	121	16	categories	category	NOUN
ijassa-348	121	17	such	such	ADJ
ijassa-348	121	18	that	that	SCONJ
ijassa-348	121	19	samples	sample	NOUN
ijassa-348	121	20	in	in	ADP
ijassa-348	121	21	one	one	NUM
ijassa-348	121	22	class	class	NOUN
ijassa-348	121	23	fall	fall	VERB
ijassa-348	121	24	entirely	entirely	ADV
ijassa-348	121	25	on	on	ADP
ijassa-348	121	26	one	one	NUM
ijassa-348	121	27	side	side	NOUN
ijassa-348	121	28	of	of	ADP
ijassa-348	121	29	h.	h.	NOUN
ijassa-348	121	30	as	as	SCONJ
ijassa-348	121	31	we	we	PRON
ijassa-348	121	32	can	can	AUX
ijassa-348	121	33	determine	determine	VERB
ijassa-348	121	34	unlimited	unlimited	ADJ
ijassa-348	121	35	number	number	NOUN
ijassa-348	121	36	of	of	ADP
ijassa-348	121	37	candidate	candidate	NOUN
ijassa-348	121	38	hyperplanes	hyperplane	NOUN
ijassa-348	121	39	,	,	PUNCT
ijassa-348	121	40	svm	svm	ADJ
ijassa-348	121	41	selects	select	NOUN
ijassa-348	121	42	only	only	ADV
ijassa-348	121	43	the	the	DET
ijassa-348	121	44	hyperplane	hyperplane	NOUN
ijassa-348	121	45	that	that	PRON
ijassa-348	121	46	maximizes	maximize	VERB
ijassa-348	121	47	distance	distance	NOUN
ijassa-348	121	48	to	to	ADP
ijassa-348	121	49	the	the	DET
ijassa-348	121	50	closest	close	ADJ
ijassa-348	121	51	data	data	NOUN
ijassa-348	121	52	samples	sample	NOUN
ijassa-348	121	53	in	in	ADP
ijassa-348	121	54	either	either	DET
ijassa-348	121	55	class	class	NOUN
ijassa-348	121	56	.	.	PUNCT
ijassa-348	122	1	this	this	PRON
ijassa-348	122	2	is	be	AUX
ijassa-348	122	3	known	know	VERB
ijassa-348	122	4	as	as	ADP
ijassa-348	122	5	margin	margin	NOUN
ijassa-348	122	6	maximization	maximization	NOUN
ijassa-348	122	7	.	.	PUNCT
ijassa-348	123	1	the	the	DET
ijassa-348	123	2	major	major	ADJ
ijassa-348	123	3	features	feature	NOUN
ijassa-348	123	4	of	of	ADP
ijassa-348	123	5	svm	svm	PROPN
ijassa-348	123	6	are	be	AUX
ijassa-348	123	7	:	:	PUNCT
ijassa-348	123	8	•	•	NUM
ijassa-348	123	9	deals	deal	NOUN
ijassa-348	123	10	with	with	ADP
ijassa-348	123	11	very	very	ADV
ijassa-348	123	12	large	large	ADJ
ijassa-348	123	13	data	data	NOUN
ijassa-348	123	14	sets	set	NOUN
ijassa-348	123	15	efficiently	efficiently	ADV
ijassa-348	123	16	.	.	PUNCT
ijassa-348	124	1	•	•	NUM
ijassa-348	124	2	multiclass	multiclass	ADJ
ijassa-348	124	3	classification	classification	NOUN
ijassa-348	124	4	can	can	AUX
ijassa-348	124	5	be	be	AUX
ijassa-348	124	6	done	do	VERB
ijassa-348	124	7	with	with	ADP
ijassa-348	124	8	any	any	DET
ijassa-348	124	9	number	number	NOUN
ijassa-348	124	10	of	of	ADP
ijassa-348	124	11	class	class	NOUN
ijassa-348	124	12	labels	label	NOUN
ijassa-348	124	13	.	.	PUNCT
ijassa-348	125	1	•	•	NUM
ijassa-348	125	2	high	high	ADJ
ijassa-348	125	3	dimensional	dimensional	ADJ
ijassa-348	125	4	data	datum	NOUN
ijassa-348	125	5	in	in	ADP
ijassa-348	125	6	both	both	CCONJ
ijassa-348	125	7	sparse	sparse	ADJ
ijassa-348	125	8	and	and	CCONJ
ijassa-348	125	9	dense	dense	ADJ
ijassa-348	125	10	formats	format	NOUN
ijassa-348	125	11	are	be	AUX
ijassa-348	125	12	supported	support	VERB
ijassa-348	125	13	.	.	PUNCT
ijassa-348	126	1	•	•	NUM
ijassa-348	126	2	expensive	expensive	ADJ
ijassa-348	126	3	computing	computing	NOUN
ijassa-348	126	4	not	not	PART
ijassa-348	126	5	required	require	VERB
ijassa-348	126	6	.	.	PUNCT
ijassa-348	127	1	•	•	NOUN
ijassa-348	127	2	used	use	VERB
ijassa-348	127	3	in	in	ADP
ijassa-348	127	4	many	many	ADJ
ijassa-348	127	5	applications	application	NOUN
ijassa-348	127	6	like	like	ADP
ijassa-348	127	7	e	e	NOUN
ijassa-348	127	8	-	-	NOUN
ijassa-348	127	9	commerce	commerce	NOUN
ijassa-348	127	10	,	,	PUNCT
ijassa-348	127	11	text	text	NOUN
ijassa-348	127	12	classification	classification	NOUN
ijassa-348	127	13	,	,	PUNCT
ijassa-348	127	14	bioinformatics	bioinformatics	NOUN
ijassa-348	127	15	,	,	PUNCT
ijassa-348	127	16	banking	banking	NOUN
ijassa-348	127	17	and	and	CCONJ
ijassa-348	127	18	other	other	ADJ
ijassa-348	127	19	areas	area	NOUN
ijassa-348	127	20	.	.	PUNCT
ijassa-348	128	1	there	there	PRON
ijassa-348	128	2	are	be	VERB
ijassa-348	128	3	many	many	ADJ
ijassa-348	128	4	real	real	ADJ
ijassa-348	128	5	time	time	NOUN
ijassa-348	128	6	applications	application	NOUN
ijassa-348	128	7	where	where	SCONJ
ijassa-348	128	8	such	such	DET
ijassa-348	128	9	a	a	DET
ijassa-348	128	10	hyperplane	hyperplane	NOUN
ijassa-348	128	11	does	do	AUX
ijassa-348	128	12	not	not	PART
ijassa-348	128	13	exist	exist	VERB
ijassa-348	128	14	.	.	PUNCT
ijassa-348	129	1	in	in	ADP
ijassa-348	129	2	such	such	ADJ
ijassa-348	129	3	cases	case	NOUN
ijassa-348	129	4	,	,	PUNCT
ijassa-348	129	5	svm	svm	PROPN
ijassa-348	129	6	utilizes	utilize	VERB
ijassa-348	129	7	a	a	DET
ijassa-348	129	8	function	function	NOUN
ijassa-348	129	9	to	to	PART
ijassa-348	129	10	transform	transform	VERB
ijassa-348	129	11	the	the	DET
ijassa-348	129	12	data	datum	NOUN
ijassa-348	129	13	into	into	ADP
ijassa-348	129	14	a	a	DET
ijassa-348	129	15	different	different	ADJ
ijassa-348	129	16	feature	feature	NOUN
ijassa-348	129	17	space	space	NOUN
ijassa-348	129	18	such	such	ADJ
ijassa-348	129	19	that	that	SCONJ
ijassa-348	129	20	there	there	PRON
ijassa-348	129	21	is	be	VERB
ijassa-348	129	22	a	a	DET
ijassa-348	129	23	possibility	possibility	NOUN
ijassa-348	129	24	of	of	ADP
ijassa-348	129	25	separation	separation	NOUN
ijassa-348	129	26	.	.	PUNCT
ijassa-348	130	1	the	the	DET
ijassa-348	130	2	function	function	NOUN
ijassa-348	130	3	that	that	PRON
ijassa-348	130	4	performs	perform	VERB
ijassa-348	130	5	such	such	DET
ijassa-348	130	6	a	a	DET
ijassa-348	130	7	transformation	transformation	NOUN
ijassa-348	130	8	is	be	AUX
ijassa-348	130	9	called	call	VERB
ijassa-348	130	10	as	as	ADP
ijassa-348	130	11	kernel	kernel	PROPN
ijassa-348	130	12	function	function	PROPN
ijassa-348	130	13	.	.	PUNCT
ijassa-348	131	1	kernels	kernel	NOUN
ijassa-348	131	2	play	play	VERB
ijassa-348	131	3	a	a	DET
ijassa-348	131	4	major	major	ADJ
ijassa-348	131	5	role	role	NOUN
ijassa-348	131	6	in	in	ADP
ijassa-348	131	7	svm	svm	PROPN
ijassa-348	131	8	.	.	PUNCT
ijassa-348	132	1	the	the	DET
ijassa-348	132	2	different	different	ADJ
ijassa-348	132	3	kernel	kernel	NOUN
ijassa-348	132	4	functions	function	NOUN
ijassa-348	132	5	widely	widely	ADV
ijassa-348	132	6	used	use	VERB
ijassa-348	132	7	along	along	ADV
ijassa-348	132	8	with	with	ADP
ijassa-348	132	9	svm	svm	PROPN
ijassa-348	132	10	are	be	AUX
ijassa-348	132	11	[	[	X
ijassa-348	132	12	25	25	NUM
ijassa-348	132	13	]	]	PUNCT
ijassa-348	132	14	as	as	SCONJ
ijassa-348	132	15	given	give	VERB
ijassa-348	132	16	below	below	ADV
ijassa-348	132	17	:	:	PUNCT
ijassa-348	132	18	i	i	PROPN
ijassa-348	132	19	)	)	PUNCT
ijassa-348	132	20	linear	linear	ADJ
ijassa-348	132	21	kernel	kernel	NOUN
ijassa-348	132	22	:	:	PUNCT
ijassa-348	133	1	k	k	PROPN
ijassa-348	133	2	(	(	PUNCT
ijassa-348	133	3	xi	xi	PROPN
ijassa-348	133	4	,	,	PUNCT
ijassa-348	133	5	xj	xj	PROPN
ijassa-348	133	6	)	)	PUNCT
ijassa-348	133	7	=	=	SYM
ijassa-348	133	8	xixj	xixj	PROPN
ijassa-348	133	9	j	j	PROPN
ijassa-348	133	10	)	)	PUNCT
ijassa-348	133	11	polynomial	polynomial	ADJ
ijassa-348	133	12	kernel	kernel	NOUN
ijassa-348	133	13	:	:	PUNCT
ijassa-348	134	1	k	k	X
ijassa-348	134	2	(	(	PUNCT
ijassa-348	134	3	x	x	X
ijassa-348	134	4	,	,	PUNCT
ijassa-348	134	5	x	x	NOUN
ijassa-348	134	6	′	′	NUM
ijassa-348	134	7	)	)	PUNCT
ijassa-348	134	8	=	=	PUNCT
ijassa-348	135	1	(	(	PUNCT
ijassa-348	135	2	xx	xx	NUM
ijassa-348	135	3	′	′	NUM
ijassa-348	136	1	+	+	CCONJ
ijassa-348	136	2	1	1	X
ijassa-348	136	3	)	)	PUNCT
ijassa-348	136	4	d	d	X
ijassa-348	136	5	k	k	X
ijassa-348	136	6	)	)	PUNCT
ijassa-348	136	7	rbf	rbf	PROPN
ijassa-348	136	8	kernel	kernel	PROPN
ijassa-348	136	9	:	:	PUNCT
ijassa-348	137	1	k	k	X
ijassa-348	137	2	(	(	PUNCT
ijassa-348	137	3	x	x	X
ijassa-348	137	4	,	,	PUNCT
ijassa-348	137	5	x	x	NOUN
ijassa-348	137	6	′	′	NUM
ijassa-348	137	7	)	)	PUNCT
ijassa-348	137	8	=	=	NOUN
ijassa-348	137	9	exp	exp	NOUN
ijassa-348	137	10	(	(	PUNCT
ijassa-348	137	11	−γ∥	−γ∥	VERB
ijassa-348	137	12	x−	x−	PROPN
ijassa-348	137	13	x	x	PUNCT
ijassa-348	137	14	′	′	NUM
ijassa-348	137	15	∥2	∥2	NOUN
ijassa-348	137	16	)	)	PUNCT
ijassa-348	138	1	l	l	NOUN
ijassa-348	138	2	)	)	PUNCT
ijassa-348	138	3	sigmoid	sigmoid	NOUN
ijassa-348	138	4	kernel	kernel	NOUN
ijassa-348	138	5	:	:	PUNCT
ijassa-348	139	1	k	k	PROPN
ijassa-348	139	2	(	(	PUNCT
ijassa-348	139	3	xi	xi	PROPN
ijassa-348	139	4	,	,	PUNCT
ijassa-348	139	5	xj	xj	PROPN
ijassa-348	139	6	)	)	PUNCT
ijassa-348	139	7	=	=	SYM
ijassa-348	139	8	tanh	tanh	PROPN
ijassa-348	139	9	(	(	PUNCT
ijassa-348	139	10	yxi	yxi	NOUN
ijassa-348	139	11	txj	txj	NOUN
ijassa-348	139	12	+	+	CCONJ
ijassa-348	139	13	r	r	NOUN
ijassa-348	139	14	)	)	PUNCT
ijassa-348	139	15	k	k	NOUN
ijassa-348	139	16	(	(	PUNCT
ijassa-348	139	17	xi	xi	PROPN
ijassa-348	139	18	,	,	PUNCT
ijassa-348	139	19	xj	xj	PROPN
ijassa-348	139	20	)	)	PUNCT
ijassa-348	139	21	=	=	SYM
ijassa-348	139	22	tanh	tanh	PROPN
ijassa-348	139	23	(	(	PUNCT
ijassa-348	139	24	yxi	yxi	NOUN
ijassa-348	139	25	txj	txj	NOUN
ijassa-348	139	26	+	+	CCONJ
ijassa-348	139	27	r	r	NOUN
ijassa-348	139	28	)	)	PUNCT
ijassa-348	139	29	advances	advance	NOUN
ijassa-348	139	30	in	in	ADP
ijassa-348	139	31	systems	system	NOUN
ijassa-348	139	32	science	science	NOUN
ijassa-348	139	33	and	and	CCONJ
ijassa-348	139	34	application	application	NOUN
ijassa-348	139	35	(	(	PUNCT
ijassa-348	139	36	2016	2016	NUM
ijassa-348	139	37	)	)	PUNCT
ijassa-348	140	1	vol.16	vol.16	PROPN
ijassa-348	140	2	no.2	no.2	PROPN
ijassa-348	140	3	21	21	NUM
ijassa-348	140	4	svm	svm	NOUN
ijassa-348	140	5	can	can	AUX
ijassa-348	140	6	be	be	AUX
ijassa-348	140	7	extended	extend	VERB
ijassa-348	140	8	to	to	ADP
ijassa-348	140	9	multi	multi	ADJ
ijassa-348	140	10	-	-	ADJ
ijassa-348	140	11	class	class	ADJ
ijassa-348	140	12	classification	classification	NOUN
ijassa-348	140	13	,	,	PUNCT
ijassa-348	140	14	a	a	DET
ijassa-348	140	15	set	set	NOUN
ijassa-348	140	16	of	of	ADP
ijassa-348	140	17	binary	binary	ADJ
ijassa-348	140	18	classifiers	classifier	NOUN
ijassa-348	140	19	are	be	AUX
ijassa-348	140	20	trained	train	VERB
ijassa-348	140	21	one	one	NUM
ijassa-348	140	22	for	for	ADP
ijassa-348	140	23	each	each	DET
ijassa-348	140	24	class	class	NOUN
ijassa-348	140	25	depending	depend	VERB
ijassa-348	140	26	on	on	ADP
ijassa-348	140	27	the	the	DET
ijassa-348	140	28	data	datum	NOUN
ijassa-348	140	29	set	set	VERB
ijassa-348	140	30	and	and	CCONJ
ijassa-348	140	31	its	its	PRON
ijassa-348	140	32	respective	respective	ADJ
ijassa-348	140	33	class	class	NOUN
ijassa-348	140	34	labels.ie	labels.ie	PROPN
ijassa-348	140	35	.	.	PUNCT
ijassa-348	141	1	if	if	SCONJ
ijassa-348	141	2	we	we	PRON
ijassa-348	141	3	train	train	VERB
ijassa-348	141	4	the	the	DET
ijassa-348	141	5	nsl	nsl	NOUN
ijassa-348	141	6	-	-	PUNCT
ijassa-348	141	7	kdd	kdd	PROPN
ijassa-348	141	8	data	datum	NOUN
ijassa-348	141	9	set	set	NOUN
ijassa-348	141	10	,	,	PUNCT
ijassa-348	141	11	then	then	ADV
ijassa-348	141	12	let	let	VERB
ijassa-348	141	13	i=15	i=15	PRON
ijassa-348	141	14	be	be	AUX
ijassa-348	141	15	a	a	DET
ijassa-348	141	16	index	index	NOUN
ijassa-348	141	17	in	in	ADP
ijassa-348	141	18	the	the	DET
ijassa-348	141	19	set	set	ADJ
ijassa-348	141	20	s=(normal	s=(normal	PROPN
ijassa-348	141	21	,	,	PUNCT
ijassa-348	141	22	probe	probe	NOUN
ijassa-348	141	23	,	,	PUNCT
ijassa-348	141	24	dos	do	NOUN
ijassa-348	141	25	,	,	PUNCT
ijassa-348	141	26	u2r	u2r	NOUN
ijassa-348	141	27	and	and	CCONJ
ijassa-348	141	28	r2l	r2l	NOUN
ijassa-348	141	29	)	)	PUNCT
ijassa-348	141	30	and	and	CCONJ
ijassa-348	141	31	let	let	VERB
ijassa-348	141	32	bi	bi	NOUN
ijassa-348	141	33	denote	denote	VERB
ijassa-348	141	34	the	the	DET
ijassa-348	141	35	matching	match	VERB
ijassa-348	141	36	binary	binary	ADJ
ijassa-348	141	37	classifier	classifier	NOUN
ijassa-348	141	38	for	for	ADP
ijassa-348	141	39	the	the	DET
ijassa-348	141	40	target	target	NOUN
ijassa-348	141	41	set	set	VERB
ijassa-348	141	42	s.	s.	PROPN
ijassa-348	141	43	similarly	similarly	ADV
ijassa-348	141	44	if	if	SCONJ
ijassa-348	141	45	we	we	PRON
ijassa-348	141	46	train	train	VERB
ijassa-348	141	47	the	the	DET
ijassa-348	141	48	unsw	unsw	PROPN
ijassa-348	141	49	-	-	PUNCT
ijassa-348	141	50	nb	nb	NOUN
ijassa-348	141	51	dataset	dataset	NOUN
ijassa-348	141	52	,	,	PUNCT
ijassa-348	141	53	then	then	ADV
ijassa-348	141	54	let	let	VERB
ijassa-348	141	55	i=	i=	PROPN
ijassa-348	141	56	1	1	NUM
ijassa-348	141	57	·	·	PUNCT
ijassa-348	141	58	·	·	PUNCT
ijassa-348	141	59	·	·	PUNCT
ijassa-348	141	60	9	9	NUM
ijassa-348	141	61	be	be	AUX
ijassa-348	141	62	a	a	DET
ijassa-348	141	63	index	index	NOUN
ijassa-348	141	64	in	in	ADP
ijassa-348	141	65	the	the	DET
ijassa-348	141	66	set	set	ADJ
ijassa-348	141	67	s=(normal	s=(normal	PROPN
ijassa-348	141	68	,	,	PUNCT
ijassa-348	141	69	analysis	analysis	NOUN
ijassa-348	141	70	,	,	PUNCT
ijassa-348	141	71	backdoor	backdoor	NOUN
ijassa-348	141	72	,	,	PUNCT
ijassa-348	141	73	reconnaissance	reconnaissance	NOUN
ijassa-348	141	74	,	,	PUNCT
ijassa-348	141	75	exploits	exploit	NOUN
ijassa-348	141	76	,	,	PUNCT
ijassa-348	141	77	fuzzers	fuzzer	NOUN
ijassa-348	141	78	,	,	PUNCT
ijassa-348	141	79	generic	generic	ADJ
ijassa-348	141	80	,	,	PUNCT
ijassa-348	141	81	dos	do	NOUN
ijassa-348	141	82	and	and	CCONJ
ijassa-348	141	83	shellcode	shellcode	NOUN
ijassa-348	141	84	)	)	PUNCT
ijassa-348	141	85	and	and	CCONJ
ijassa-348	141	86	let	let	VERB
ijassa-348	141	87	bi	bi	NOUN
ijassa-348	141	88	denote	denote	VERB
ijassa-348	141	89	the	the	DET
ijassa-348	141	90	matching	match	VERB
ijassa-348	141	91	binary	binary	ADJ
ijassa-348	141	92	classifier	classifier	NOUN
ijassa-348	141	93	for	for	ADP
ijassa-348	141	94	the	the	DET
ijassa-348	141	95	target	target	NOUN
ijassa-348	141	96	set	set	VERB
ijassa-348	141	97	s.	s.	PROPN
ijassa-348	141	98	thus	thus	ADV
ijassa-348	141	99	the	the	DET
ijassa-348	141	100	observations	observation	NOUN
ijassa-348	141	101	are	be	AUX
ijassa-348	141	102	classified	classify	VERB
ijassa-348	141	103	using	use	VERB
ijassa-348	141	104	one	one	NUM
ijassa-348	141	105	-	-	PUNCT
ijassa-348	141	106	versus	versus	ADP
ijassa-348	141	107	-	-	PUNCT
ijassa-348	141	108	all	all	PRON
ijassa-348	141	109	approach	approach	NOUN
ijassa-348	141	110	in	in	ADP
ijassa-348	141	111	both	both	CCONJ
ijassa-348	141	112	the	the	DET
ijassa-348	141	113	datasets	dataset	NOUN
ijassa-348	141	114	.	.	PUNCT
ijassa-348	142	1	to	to	PART
ijassa-348	142	2	distinguish	distinguish	VERB
ijassa-348	142	3	among	among	ADP
ijassa-348	142	4	the	the	DET
ijassa-348	142	5	binary	binary	ADJ
ijassa-348	142	6	classifiers	classifier	NOUN
ijassa-348	142	7	,	,	PUNCT
ijassa-348	142	8	we	we	PRON
ijassa-348	142	9	deploy	deploy	VERB
ijassa-348	142	10	manager	manager	NOUN
ijassa-348	142	11	to	to	PART
ijassa-348	142	12	denote	denote	VERB
ijassa-348	142	13	one	one	NUM
ijassa-348	142	14	set	set	NOUN
ijassa-348	142	15	of	of	ADP
ijassa-348	142	16	classification	classification	NOUN
ijassa-348	142	17	.	.	PUNCT
ijassa-348	143	1	svm	svm	PROPN
ijassa-348	143	2	produces	produce	VERB
ijassa-348	143	3	the	the	DET
ijassa-348	143	4	best	good	ADJ
ijassa-348	143	5	results	result	NOUN
ijassa-348	143	6	when	when	SCONJ
ijassa-348	143	7	the	the	DET
ijassa-348	143	8	rbf	rbf	PROPN
ijassa-348	143	9	kernel	kernel	PROPN
ijassa-348	143	10	function	function	PROPN
ijassa-348	143	11	is	be	AUX
ijassa-348	143	12	utilized	utilize	VERB
ijassa-348	143	13	.	.	PUNCT
ijassa-348	144	1	experimental	experimental	ADJ
ijassa-348	144	2	results	result	NOUN
ijassa-348	144	3	show	show	VERB
ijassa-348	144	4	that	that	SCONJ
ijassa-348	144	5	the	the	DET
ijassa-348	144	6	performance	performance	NOUN
ijassa-348	144	7	of	of	ADP
ijassa-348	144	8	svm	svm	ADJ
ijassa-348	144	9	classifiers	classifier	NOUN
ijassa-348	144	10	will	will	AUX
ijassa-348	144	11	differ	differ	VERB
ijassa-348	144	12	with	with	ADP
ijassa-348	144	13	the	the	DET
ijassa-348	144	14	selection	selection	NOUN
ijassa-348	144	15	of	of	ADP
ijassa-348	144	16	rbf	rbf	PROPN
ijassa-348	144	17	function	function	PROPN
ijassa-348	144	18	.	.	PUNCT
ijassa-348	145	1	therefore	therefore	ADV
ijassa-348	145	2	in	in	ADP
ijassa-348	145	3	this	this	DET
ijassa-348	145	4	paper	paper	NOUN
ijassa-348	145	5	we	we	PRON
ijassa-348	145	6	train	train	VERB
ijassa-348	145	7	the	the	DET
ijassa-348	145	8	svm	svm	ADJ
ijassa-348	145	9	manager	manager	NOUN
ijassa-348	145	10	with	with	ADP
ijassa-348	145	11	five	five	NUM
ijassa-348	145	12	different	different	ADJ
ijassa-348	145	13	rbf	rbf	PROPN
ijassa-348	145	14	values=	values=	NUM
ijassa-348	145	15	[	[	PUNCT
ijassa-348	145	16	5	5	NUM
ijassa-348	145	17	,	,	PUNCT
ijassa-348	145	18	2	2	NUM
ijassa-348	145	19	,	,	PUNCT
ijassa-348	145	20	1	1	NUM
ijassa-348	145	21	,	,	PUNCT
ijassa-348	145	22	0.5	0.5	NUM
ijassa-348	145	23	,	,	PUNCT
ijassa-348	145	24	0.1	0.1	NUM
ijassa-348	145	25	]	]	PUNCT
ijassa-348	145	26	for	for	SCONJ
ijassa-348	145	27	both	both	DET
ijassa-348	145	28	datasets	dataset	NOUN
ijassa-348	145	29	to	to	PART
ijassa-348	145	30	ensure	ensure	VERB
ijassa-348	145	31	that	that	SCONJ
ijassa-348	145	32	the	the	DET
ijassa-348	145	33	svm	svm	ADJ
ijassa-348	145	34	algorithm	algorithm	NOUN
ijassa-348	145	35	is	be	AUX
ijassa-348	145	36	utilized	utilize	VERB
ijassa-348	145	37	maximally	maximally	ADV
ijassa-348	145	38	.	.	PUNCT
ijassa-348	146	1	this	this	DET
ijassa-348	146	2	approach	approach	NOUN
ijassa-348	146	3	will	will	AUX
ijassa-348	146	4	ensure	ensure	VERB
ijassa-348	146	5	greater	great	ADJ
ijassa-348	146	6	diversity	diversity	NOUN
ijassa-348	146	7	of	of	ADP
ijassa-348	146	8	managers	manager	NOUN
ijassa-348	146	9	in	in	ADP
ijassa-348	146	10	ensemble	ensemble	ADJ
ijassa-348	146	11	classifier	classifier	NOUN
ijassa-348	146	12	as	as	SCONJ
ijassa-348	146	13	the	the	DET
ijassa-348	146	14	accuracy	accuracy	NOUN
ijassa-348	146	15	will	will	AUX
ijassa-348	146	16	vary	vary	VERB
ijassa-348	146	17	for	for	ADP
ijassa-348	146	18	each	each	DET
ijassa-348	146	19	binary	binary	ADJ
ijassa-348	146	20	classifier	classifier	NOUN
ijassa-348	146	21	according	accord	VERB
ijassa-348	146	22	to	to	ADP
ijassa-348	146	23	the	the	DET
ijassa-348	146	24	selected	select	VERB
ijassa-348	146	25	rbf	rbf	PROPN
ijassa-348	146	26	values	value	NOUN
ijassa-348	146	27	in	in	ADP
ijassa-348	146	28	the	the	DET
ijassa-348	146	29	vector	vector	NOUN
ijassa-348	146	30	.	.	PUNCT
ijassa-348	147	1	construct	construct	VERB
ijassa-348	147	2	a	a	DET
ijassa-348	147	3	set	set	NOUN
ijassa-348	147	4	of	of	ADP
ijassa-348	147	5	binary	binary	ADJ
ijassa-348	147	6	classifiers	classifier	NOUN
ijassa-348	147	7	f	f	PROPN
ijassa-348	147	8	1	1	NUM
ijassa-348	147	9	,	,	PUNCT
ijassa-348	147	10	f	f	PROPN
ijassa-348	147	11	2	2	NUM
ijassa-348	147	12	·	·	PUNCT
ijassa-348	147	13	·	·	PUNCT
ijassa-348	147	14	·	·	PUNCT
ijassa-348	148	1	f	f	X
ijassa-348	148	2	n	n	ADV
ijassa-348	148	3	for	for	ADP
ijassa-348	148	4	1	1	NUM
ijassa-348	148	5	·	·	PUNCT
ijassa-348	148	6	·	·	PUNCT
ijassa-348	148	7	·	·	PUNCT
ijassa-348	148	8	n	n	NUM
ijassa-348	148	9	classes	class	NOUN
ijassa-348	148	10	each	each	PRON
ijassa-348	148	11	trained	train	VERB
ijassa-348	148	12	to	to	PART
ijassa-348	148	13	differentiate	differentiate	VERB
ijassa-348	148	14	one	one	NUM
ijassa-348	148	15	class	class	NOUN
ijassa-348	148	16	from	from	ADP
ijassa-348	148	17	the	the	DET
ijassa-348	148	18	rest	rest	NOUN
ijassa-348	148	19	.	.	PUNCT
ijassa-348	149	1	a	a	DET
ijassa-348	149	2	multi	multi	ADJ
ijassa-348	149	3	class	class	NOUN
ijassa-348	149	4	categorization	categorization	NOUN
ijassa-348	149	5	can	can	AUX
ijassa-348	149	6	be	be	AUX
ijassa-348	149	7	obtained	obtain	VERB
ijassa-348	149	8	by	by	ADP
ijassa-348	149	9	combining	combine	VERB
ijassa-348	149	10	them	they	PRON
ijassa-348	149	11	according	accord	VERB
ijassa-348	149	12	to	to	ADP
ijassa-348	149	13	the	the	DET
ijassa-348	149	14	maximal	maximal	ADJ
ijassa-348	149	15	output	output	NOUN
ijassa-348	149	16	before	before	ADP
ijassa-348	149	17	applying	apply	VERB
ijassa-348	149	18	the	the	DET
ijassa-348	149	19	sgn	sgn	NOUN
ijassa-348	149	20	function	function	NOUN
ijassa-348	149	21	.	.	PUNCT
ijassa-348	150	1	argmax	argmax	PROPN
ijassa-348	150	2	gk(x	gk(x	PROPN
ijassa-348	150	3	)	)	PUNCT
ijassa-348	150	4	where	where	SCONJ
ijassa-348	150	5	gk(x	gk(x	X
ijassa-348	150	6	)	)	PUNCT
ijassa-348	151	1	=	=	SYM
ijassa-348	151	2	n∑	n∑	PROPN
ijassa-348	151	3	i=1	i=1	PROPN
ijassa-348	152	1	yiai	yiai	PROPN
ijassa-348	152	2	kk(x	kk(x	PROPN
ijassa-348	152	3	,	,	PUNCT
ijassa-348	152	4	xi	xi	X
ijassa-348	152	5	)	)	PUNCT
ijassa-348	153	1	+	+	CCONJ
ijassa-348	153	2	bk	bk	VERB
ijassa-348	153	3	where	where	SCONJ
ijassa-348	153	4	k	k	PROPN
ijassa-348	153	5	=	=	SYM
ijassa-348	153	6	1	1	NUM
ijassa-348	153	7	·	·	PUNCT
ijassa-348	153	8	·	·	PUNCT
ijassa-348	153	9	·	·	PUNCT
ijassa-348	153	10	n.	n.	NOUN
ijassa-348	153	11	(	(	PUNCT
ijassa-348	153	12	2	2	NUM
ijassa-348	153	13	)	)	PUNCT
ijassa-348	153	14	wherein	wherein	SCONJ
ijassa-348	153	15	gk	gk	PROPN
ijassa-348	153	16	(	(	PUNCT
ijassa-348	153	17	x	x	NOUN
ijassa-348	153	18	)	)	PUNCT
ijassa-348	153	19	returns	return	VERB
ijassa-348	153	20	a	a	DET
ijassa-348	153	21	signed	sign	VERB
ijassa-348	153	22	real	real	ADJ
ijassa-348	153	23	value	value	NOUN
ijassa-348	153	24	which	which	PRON
ijassa-348	153	25	is	be	AUX
ijassa-348	153	26	the	the	DET
ijassa-348	153	27	distance	distance	NOUN
ijassa-348	153	28	from	from	ADP
ijassa-348	153	29	the	the	DET
ijassa-348	153	30	hyper	hyper	ADJ
ijassa-348	153	31	plane	plane	NOUN
ijassa-348	153	32	to	to	ADP
ijassa-348	153	33	the	the	DET
ijassa-348	153	34	point	point	NOUN
ijassa-348	153	35	x.	x.	NOUN
ijassa-348	154	1	this	this	DET
ijassa-348	154	2	value	value	NOUN
ijassa-348	154	3	is	be	AUX
ijassa-348	154	4	referred	refer	VERB
ijassa-348	154	5	as	as	ADP
ijassa-348	154	6	the	the	DET
ijassa-348	154	7	confidence	confidence	NOUN
ijassa-348	154	8	value	value	NOUN
ijassa-348	154	9	.	.	PUNCT
ijassa-348	155	1	the	the	PRON
ijassa-348	155	2	higher	high	ADJ
ijassa-348	155	3	the	the	DET
ijassa-348	155	4	value	value	NOUN
ijassa-348	155	5	,	,	PUNCT
ijassa-348	155	6	the	the	DET
ijassa-348	155	7	more	more	ADJ
ijassa-348	155	8	is	be	AUX
ijassa-348	155	9	the	the	DET
ijassa-348	155	10	confident	confident	ADJ
ijassa-348	155	11	that	that	SCONJ
ijassa-348	155	12	the	the	DET
ijassa-348	155	13	point	point	NOUN
ijassa-348	155	14	x	x	PUNCT
ijassa-348	155	15	belongs	belong	VERB
ijassa-348	155	16	to	to	ADP
ijassa-348	155	17	positive	positive	ADJ
ijassa-348	155	18	class	class	NOUN
ijassa-348	155	19	.	.	PUNCT
ijassa-348	156	1	hence	hence	ADV
ijassa-348	156	2	we	we	PRON
ijassa-348	156	3	need	need	VERB
ijassa-348	156	4	to	to	PART
ijassa-348	156	5	assign	assign	VERB
ijassa-348	156	6	x	x	PUNCT
ijassa-348	156	7	to	to	ADP
ijassa-348	156	8	the	the	DET
ijassa-348	156	9	class	class	NOUN
ijassa-348	156	10	having	have	VERB
ijassa-348	156	11	highest	high	ADJ
ijassa-348	156	12	confidence	confidence	NOUN
ijassa-348	156	13	value	value	NOUN
ijassa-348	156	14	.	.	PUNCT
ijassa-348	157	1	given	give	VERB
ijassa-348	157	2	normal	normal	ADJ
ijassa-348	157	3	data	datum	NOUN
ijassa-348	157	4	χ	χ	NOUN
ijassa-348	157	5	=	=	PUNCT
ijassa-348	157	6	{	{	PUNCT
ijassa-348	157	7	x1	x1	PROPN
ijassa-348	157	8	,	,	PUNCT
ijassa-348	157	9	x2	x2	PROPN
ijassa-348	157	10	,	,	PUNCT
ijassa-348	157	11	...	...	PUNCT
ijassa-348	157	12	xm	xm	X
ijassa-348	157	13	}	}	PUNCT
ijassa-348	157	14	∈	∈	PROPN
ijassa-348	157	15	rd	rd	NOUN
ijassa-348	157	16	and	and	CCONJ
ijassa-348	157	17	let	let	VERB
ijassa-348	157	18	r	r	NOUN
ijassa-348	157	19	be	be	AUX
ijassa-348	157	20	the	the	DET
ijassa-348	157	21	radius	radius	NOUN
ijassa-348	157	22	of	of	ADP
ijassa-348	157	23	the	the	DET
ijassa-348	157	24	hypersphere	hypersphere	NOUN
ijassa-348	157	25	and	and	CCONJ
ijassa-348	157	26	c	c	PROPN
ijassa-348	157	27	∈	∈	PROPN
ijassa-348	157	28	rd	rd	PROPN
ijassa-348	157	29	which	which	PRON
ijassa-348	157	30	is	be	AUX
ijassa-348	157	31	the	the	DET
ijassa-348	157	32	center	center	NOUN
ijassa-348	157	33	.	.	PUNCT
ijassa-348	158	1	the	the	DET
ijassa-348	158	2	optimization	optimization	NOUN
ijassa-348	158	3	problem	problem	NOUN
ijassa-348	158	4	can	can	AUX
ijassa-348	158	5	be	be	AUX
ijassa-348	158	6	solved	solve	VERB
ijassa-348	158	7	by	by	ADP
ijassa-348	158	8	determining	determine	VERB
ijassa-348	158	9	the	the	DET
ijassa-348	158	10	minimum	minimum	NOUN
ijassa-348	158	11	enclosing	enclosing	NOUN
ijassa-348	158	12	hypersphere	hypersphere	X
ijassa-348	158	13	.	.	PUNCT
ijassa-348	159	1	minimize	minimize	VERB
ijassa-348	159	2	r2	r2	PROPN
ijassa-348	159	3	subject	subject	ADJ
ijassa-348	159	4	to	to	ADP
ijassa-348	159	5	∥	∥	PROPN
ijassa-348	159	6	ϕ	ϕ	X
ijassa-348	159	7	(	(	PUNCT
ijassa-348	159	8	xj)−	xj)−	PROPN
ijassa-348	159	9	c	c	PROPN
ijassa-348	159	10	∥2	∥2	NOUN
ijassa-348	159	11	≤	≤	NUM
ijassa-348	159	12	r2	r2	NOUN
ijassa-348	159	13	,	,	PUNCT
ijassa-348	159	14	j	j	PROPN
ijassa-348	159	15	=	=	SYM
ijassa-348	159	16	1	1	NUM
ijassa-348	159	17	,	,	PUNCT
ijassa-348	159	18	...	...	PUNCT
ijassa-348	159	19	m	m	VERB
ijassa-348	159	20	(	(	PUNCT
ijassa-348	159	21	3	3	NUM
ijassa-348	159	22	)	)	PUNCT
ijassa-348	159	23	l	l	NOUN
ijassa-348	159	24	(	(	PUNCT
ijassa-348	159	25	c	c	X
ijassa-348	159	26	,	,	PUNCT
ijassa-348	159	27	r	r	NOUN
ijassa-348	159	28	,	,	PUNCT
ijassa-348	159	29	α	α	NOUN
ijassa-348	159	30	)	)	PUNCT
ijassa-348	159	31	=	=	SYM
ijassa-348	159	32	r2	r2	PROPN
ijassa-348	159	33	+	+	CCONJ
ijassa-348	159	34	m∑	m∑	ADV
ijassa-348	159	35	j=1	j=1	NOUN
ijassa-348	159	36	αj{∥	αj{∥	NOUN
ijassa-348	159	37	ϕ	ϕ	X
ijassa-348	159	38	(	(	PUNCT
ijassa-348	159	39	xi	xi	PROPN
ijassa-348	159	40	−	−	PROPN
ijassa-348	159	41	c	c	X
ijassa-348	159	42	)	)	PUNCT
ijassa-348	159	43	∥2	∥2	NOUN
ijassa-348	160	1	−	−	NOUN
ijassa-348	160	2	r2	r2	PROPN
ijassa-348	160	3	}	}	PUNCT
ijassa-348	160	4	(	(	PUNCT
ijassa-348	160	5	4	4	X
ijassa-348	160	6	)	)	PUNCT
ijassa-348	160	7	22	22	NUM
ijassa-348	160	8	sumaiya	sumaiya	NOUN
ijassa-348	160	9	thaseen	thaseen	PROPN
ijassa-348	160	10	and	and	CCONJ
ijassa-348	160	11	ch.aswani	ch.aswani	X
ijassa-348	161	1	kumar	kumar	PROPN
ijassa-348	161	2	:	:	PUNCT
ijassa-348	161	3	intrusion	intrusion	NOUN
ijassa-348	161	4	detection	detection	NOUN
ijassa-348	161	5	model	model	NOUN
ijassa-348	161	6	using	use	VERB
ijassa-348	161	7	pca	pca	PROPN
ijassa-348	161	8	and	and	CCONJ
ijassa-348	161	9	...	...	PUNCT
ijassa-348	161	10	setting	set	VERB
ijassa-348	161	11	the	the	DET
ijassa-348	161	12	derivatives	derivative	NOUN
ijassa-348	161	13	δl	δl	X
ijassa-348	161	14	(	(	PUNCT
ijassa-348	161	15	c	c	X
ijassa-348	161	16	,	,	PUNCT
ijassa-348	161	17	r	r	NOUN
ijassa-348	161	18	,	,	PUNCT
ijassa-348	161	19	α	α	NOUN
ijassa-348	161	20	)	)	PUNCT
ijassa-348	161	21	δc	δc	NOUN
ijassa-348	161	22	=	=	SYM
ijassa-348	161	23	c	c	NOUN
ijassa-348	161	24	m∑	m∑	X
ijassa-348	161	25	j=1	j=1	ADJ
ijassa-348	161	26	αj	αj	PROPN
ijassa-348	161	27	(	(	PUNCT
ijassa-348	161	28	ϕ	ϕ	X
ijassa-348	161	29	(	(	PUNCT
ijassa-348	161	30	xj)−	xj)−	NOUN
ijassa-348	161	31	c	c	X
ijassa-348	161	32	)	)	PUNCT
ijassa-348	161	33	=	=	SYM
ijassa-348	161	34	0	0	PUNCT
ijassa-348	162	1	(	(	PUNCT
ijassa-348	162	2	5	5	X
ijassa-348	162	3	)	)	PUNCT
ijassa-348	162	4	we	we	PRON
ijassa-348	162	5	can	can	AUX
ijassa-348	162	6	obtain	obtain	VERB
ijassa-348	162	7	the	the	DET
ijassa-348	162	8	following	follow	VERB
ijassa-348	162	9	equation,∑m	equation,∑m	NOUN
ijassa-348	162	10	j=1	j=1	NOUN
ijassa-348	162	11	αj	αj	X
ijassa-348	162	12	=	=	SYM
ijassa-348	162	13	1	1	NUM
ijassa-348	162	14	and	and	CCONJ
ijassa-348	162	15	c	c	NOUN
ijassa-348	162	16	=	=	SYM
ijassa-348	162	17	∑m	∑m	PROPN
ijassa-348	162	18	j=1	j=1	PROPN
ijassa-348	162	19	αjϕ	αjϕ	PROPN
ijassa-348	162	20	(	(	PUNCT
ijassa-348	162	21	xj	xj	NOUN
ijassa-348	162	22	)	)	PUNCT
ijassa-348	162	23	hence	hence	ADV
ijassa-348	162	24	the	the	DET
ijassa-348	162	25	equation	equation	NOUN
ijassa-348	162	26	(	(	PUNCT
ijassa-348	162	27	4	4	X
ijassa-348	162	28	)	)	PUNCT
ijassa-348	162	29	becomes	become	VERB
ijassa-348	162	30	,	,	PUNCT
ijassa-348	162	31	l	l	NOUN
ijassa-348	162	32	(	(	PUNCT
ijassa-348	162	33	c	c	X
ijassa-348	162	34	,	,	PUNCT
ijassa-348	162	35	r	r	NOUN
ijassa-348	162	36	,	,	PUNCT
ijassa-348	162	37	α	α	NOUN
ijassa-348	162	38	)	)	PUNCT
ijassa-348	162	39	=	=	PUNCT
ijassa-348	163	1	m∑	m∑	CCONJ
ijassa-348	163	2	j=1	j=1	ADJ
ijassa-348	163	3	αjk	αjk	PROPN
ijassa-348	163	4	(	(	PUNCT
ijassa-348	163	5	xj	xj	PROPN
ijassa-348	163	6	,	,	PUNCT
ijassa-348	163	7	xj)−	xj)−	PROPN
ijassa-348	163	8	m∑	m∑	INTJ
ijassa-348	163	9	i	i	PRON
ijassa-348	163	10	,	,	PUNCT
ijassa-348	163	11	j=1	j=1	PROPN
ijassa-348	163	12	αiαjk	αiαjk	NOUN
ijassa-348	163	13	(	(	PUNCT
ijassa-348	163	14	xi	xi	PROPN
ijassa-348	163	15	,	,	PUNCT
ijassa-348	163	16	xj	xj	PROPN
ijassa-348	163	17	)	)	PUNCT
ijassa-348	163	18	(	(	PUNCT
ijassa-348	163	19	6	6	NUM
ijassa-348	163	20	)	)	PUNCT
ijassa-348	163	21	which	which	PRON
ijassa-348	163	22	is	be	AUX
ijassa-348	163	23	the	the	DET
ijassa-348	163	24	dual	dual	ADJ
ijassa-348	163	25	form	form	NOUN
ijassa-348	163	26	of	of	ADP
ijassa-348	163	27	equation	equation	NOUN
ijassa-348	163	28	(	(	PUNCT
ijassa-348	163	29	4	4	NUM
ijassa-348	163	30	)	)	PUNCT
ijassa-348	163	31	.	.	PUNCT
ijassa-348	164	1	the	the	DET
ijassa-348	164	2	dual	dual	ADJ
ijassa-348	164	3	form	form	NOUN
ijassa-348	164	4	of	of	ADP
ijassa-348	164	5	α	α	PRON
ijassa-348	164	6	can	can	AUX
ijassa-348	164	7	be	be	AUX
ijassa-348	164	8	obtained	obtain	VERB
ijassa-348	164	9	by	by	ADP
ijassa-348	164	10	solving	solve	VERB
ijassa-348	164	11	the	the	DET
ijassa-348	164	12	optimization	optimization	NOUN
ijassa-348	164	13	problem	problem	NOUN
ijassa-348	164	14	,	,	PUNCT
ijassa-348	164	15	maximizing	maximizing	NOUN
ijassa-348	164	16	,	,	PUNCT
ijassa-348	164	17	w	w	PROPN
ijassa-348	164	18	(	(	PUNCT
ijassa-348	164	19	α	α	NOUN
ijassa-348	164	20	)	)	PUNCT
ijassa-348	164	21	=	=	PUNCT
ijassa-348	165	1	m∑	m∑	CCONJ
ijassa-348	165	2	i=1	i=1	PROPN
ijassa-348	165	3	αik	αik	PROPN
ijassa-348	165	4	(	(	PUNCT
ijassa-348	165	5	xi	xi	PROPN
ijassa-348	165	6	,	,	PUNCT
ijassa-348	165	7	xi)−	xi)−	PROPN
ijassa-348	166	1	m∑	m∑	INTJ
ijassa-348	166	2	i	i	PRON
ijassa-348	166	3	,	,	PUNCT
ijassa-348	166	4	j=1	j=1	PROPN
ijassa-348	166	5	αiαjk	αiαjk	NOUN
ijassa-348	166	6	(	(	PUNCT
ijassa-348	166	7	xi	xi	PROPN
ijassa-348	166	8	,	,	PUNCT
ijassa-348	166	9	xj	xj	PROPN
ijassa-348	166	10	)	)	PUNCT
ijassa-348	166	11	(	(	PUNCT
ijassa-348	166	12	7	7	X
ijassa-348	166	13	)	)	PUNCT
ijassa-348	166	14	subject	subject	ADJ
ijassa-348	166	15	to∑m	to∑m	PROPN
ijassa-348	166	16	i=1	i=1	PROPN
ijassa-348	167	1	=	=	PUNCT
ijassa-348	167	2	αi	αi	NOUN
ijassa-348	167	3	=	=	SYM
ijassa-348	167	4	1	1	NUM
ijassa-348	167	5	and	and	CCONJ
ijassa-348	167	6	αi	αi	PRON
ijassa-348	167	7	≥	≥	NOUN
ijassa-348	167	8	0	0	NUM
ijassa-348	167	9	,	,	PUNCT
ijassa-348	167	10	i	i	PRON
ijassa-348	167	11	=	=	NOUN
ijassa-348	167	12	1	1	NUM
ijassa-348	167	13	to	to	PART
ijassa-348	167	14	m	m	PRON
ijassa-348	167	15	it	it	PRON
ijassa-348	167	16	should	should	AUX
ijassa-348	167	17	be	be	AUX
ijassa-348	167	18	noted	note	VERB
ijassa-348	167	19	that	that	SCONJ
ijassa-348	167	20	lagrange	lagrange	NOUN
ijassa-348	167	21	multiplier	multiplier	ADV
ijassa-348	167	22	can	can	AUX
ijassa-348	167	23	be	be	AUX
ijassa-348	167	24	non	non	ADJ
ijassa-348	167	25	-	-	ADJ
ijassa-348	167	26	zero	zero	NUM
ijassa-348	167	27	only	only	ADV
ijassa-348	167	28	if	if	SCONJ
ijassa-348	167	29	the	the	DET
ijassa-348	167	30	inequality	inequality	NOUN
ijassa-348	167	31	constraint	constraint	NOUN
ijassa-348	167	32	is	be	AUX
ijassa-348	167	33	an	an	DET
ijassa-348	167	34	equality	equality	NOUN
ijassa-348	167	35	for	for	ADP
ijassa-348	167	36	the	the	DET
ijassa-348	167	37	solution	solution	NOUN
ijassa-348	167	38	.	.	PUNCT
ijassa-348	168	1	the	the	DET
ijassa-348	168	2	complementarity	complementarity	NOUN
ijassa-348	168	3	conditions	condition	NOUN
ijassa-348	168	4	are	be	AUX
ijassa-348	168	5	satisfied	satisfied	ADJ
ijassa-348	168	6	by	by	ADP
ijassa-348	168	7	the	the	DET
ijassa-348	168	8	optimal	optimal	ADJ
ijassa-348	168	9	solutions	solution	NOUN
ijassa-348	168	10	α,(c	α,(c	NOUN
ijassa-348	168	11	,	,	PUNCT
ijassa-348	168	12	γ	γ	NOUN
ijassa-348	168	13	)	)	PUNCT
ijassa-348	168	14	αi{∥	αi{∥	NUM
ijassa-348	168	15	ϕ	ϕ	NOUN
ijassa-348	168	16	(	(	PUNCT
ijassa-348	169	1	xi)−	xi)−	PROPN
ijassa-348	169	2	c	c	NOUN
ijassa-348	169	3	∥2	∥2	PUNCT
ijassa-348	169	4	−	−	NOUN
ijassa-348	169	5	r2	r2	PROPN
ijassa-348	169	6	}	}	PUNCT
ijassa-348	169	7	,	,	PUNCT
ijassa-348	169	8	i	i	PRON
ijassa-348	169	9	=	=	NOUN
ijassa-348	170	1	1	1	NUM
ijassa-348	170	2	...	...	SYM
ijassa-348	170	3	m	m	VERB
ijassa-348	170	4	(	(	PUNCT
ijassa-348	170	5	8)	8)	NUM
ijassa-348	170	6	hence	hence	ADV
ijassa-348	170	7	it	it	PRON
ijassa-348	170	8	implies	imply	VERB
ijassa-348	170	9	that	that	SCONJ
ijassa-348	170	10	the	the	DET
ijassa-348	170	11	training	training	NOUN
ijassa-348	170	12	samples	sample	NOUN
ijassa-348	170	13	x	x	X
ijassa-348	170	14	i	i	PRON
ijassa-348	170	15	lie	lie	VERB
ijassa-348	170	16	on	on	ADP
ijassa-348	170	17	the	the	DET
ijassa-348	170	18	surface	surface	NOUN
ijassa-348	170	19	of	of	ADP
ijassa-348	170	20	the	the	DET
ijassa-348	170	21	optimal	optimal	ADJ
ijassa-348	170	22	hypersphere	hypersphere	NOUN
ijassa-348	170	23	corresponding	correspond	VERB
ijassa-348	170	24	to	to	ADP
ijassa-348	170	25	αi	αi	PRON
ijassa-348	170	26	>	>	X
ijassa-348	171	1	0	0	X
ijassa-348	171	2	.	.	PUNCT
ijassa-348	172	1	the	the	DET
ijassa-348	172	2	decision	decision	NOUN
ijassa-348	172	3	function	function	NOUN
ijassa-348	172	4	becomes	become	VERB
ijassa-348	172	5	,	,	PUNCT
ijassa-348	172	6	f	f	PROPN
ijassa-348	172	7	(	(	PUNCT
ijassa-348	172	8	x	x	X
ijassa-348	172	9	)	)	PUNCT
ijassa-348	172	10	=	=	SYM
ijassa-348	172	11	sgn	sgn	NOUN
ijassa-348	172	12	(	(	PUNCT
ijassa-348	172	13	r2	r2	PROPN
ijassa-348	173	1	−	−	PROPN
ijassa-348	173	2	∥	∥	NUM
ijassa-348	173	3	ϕ	ϕ	NOUN
ijassa-348	173	4	(	(	PUNCT
ijassa-348	173	5	x)−	x)−	PROPN
ijassa-348	173	6	c	c	PROPN
ijassa-348	173	7	∥2	∥2	PROPN
ijassa-348	173	8	)	)	PUNCT
ijassa-348	174	1	this	this	PRON
ijassa-348	174	2	implies	imply	VERB
ijassa-348	174	3	,	,	PUNCT
ijassa-348	174	4	=	=	SYM
ijassa-348	174	5	sgn(r2	sgn(r2	NOUN
ijassa-348	174	6	−	−	PROPN
ijassa-348	174	7	ϕ(x).ϕ(x)−	ϕ(x).ϕ(x)−	PROPN
ijassa-348	174	8	2	2	NUM
ijassa-348	174	9	m∑	m∑	NOUN
ijassa-348	174	10	i=1	i=1	PROPN
ijassa-348	174	11	αiϕ(x).ϕ(xi	αiϕ(x).ϕ(xi	PROPN
ijassa-348	174	12	)	)	PUNCT
ijassa-348	175	1	+	+	CCONJ
ijassa-348	175	2	m∑	m∑	INTJ
ijassa-348	175	3	i	i	PRON
ijassa-348	175	4	,	,	PUNCT
ijassa-348	175	5	j=1	j=1	PROPN
ijassa-348	175	6	αiαj(ϕ(xi).ϕ(xj	αiαj(ϕ(xi).ϕ(xj	NOUN
ijassa-348	175	7	)	)	PUNCT
ijassa-348	175	8	)	)	PUNCT
ijassa-348	175	9	)	)	PUNCT
ijassa-348	176	1	=	=	PUNCT
ijassa-348	176	2	sgn(r2	sgn(r2	NOUN
ijassa-348	176	3	−	−	PROPN
ijassa-348	176	4	k(x	k(x	PROPN
ijassa-348	176	5	,	,	PUNCT
ijassa-348	176	6	x)−	x)−	PROPN
ijassa-348	176	7	2	2	NUM
ijassa-348	176	8	m∑	m∑	VERB
ijassa-348	176	9	i=1	i=1	PROPN
ijassa-348	176	10	ϕ(xi)k(x	ϕ(xi)k(x	PROPN
ijassa-348	176	11	,	,	PUNCT
ijassa-348	176	12	xi	xi	ADJ
ijassa-348	176	13	)	)	PUNCT
ijassa-348	177	1	+	+	CCONJ
ijassa-348	177	2	m∑	m∑	INTJ
ijassa-348	177	3	i	i	PRON
ijassa-348	177	4	,	,	PUNCT
ijassa-348	177	5	j=1	j=1	PROPN
ijassa-348	177	6	ϕiϕjk(xi	ϕiϕjk(xi	PROPN
ijassa-348	177	7	,	,	PUNCT
ijassa-348	177	8	xj	xj	PROPN
ijassa-348	177	9	)	)	PUNCT
ijassa-348	177	10	)	)	PUNCT
ijassa-348	177	11	)	)	PUNCT
ijassa-348	177	12	)	)	PUNCT
ijassa-348	178	1	(	(	PUNCT
ijassa-348	178	2	9	9	X
ijassa-348	178	3	)	)	PUNCT
ijassa-348	178	4	thus	thus	ADV
ijassa-348	178	5	the	the	DET
ijassa-348	178	6	aim	aim	NOUN
ijassa-348	178	7	of	of	ADP
ijassa-348	178	8	obtaining	obtain	VERB
ijassa-348	178	9	minimum	minimum	ADJ
ijassa-348	178	10	enclosing	enclosing	NOUN
ijassa-348	178	11	hypersphere	hypersphere	NOUN
ijassa-348	178	12	containing	contain	VERB
ijassa-348	178	13	all	all	DET
ijassa-348	178	14	training	training	NOUN
ijassa-348	178	15	samples	sample	NOUN
ijassa-348	178	16	is	be	AUX
ijassa-348	178	17	satisfied	satisfied	ADJ
ijassa-348	178	18	.	.	PUNCT
ijassa-348	179	1	advances	advance	NOUN
ijassa-348	179	2	in	in	ADP
ijassa-348	179	3	systems	system	NOUN
ijassa-348	179	4	science	science	NOUN
ijassa-348	179	5	and	and	CCONJ
ijassa-348	179	6	application	application	NOUN
ijassa-348	179	7	(	(	PUNCT
ijassa-348	179	8	2016	2016	NUM
ijassa-348	179	9	)	)	PUNCT
ijassa-348	180	1	vol.16	vol.16	PROPN
ijassa-348	180	2	no.2	no.2	PROPN
ijassa-348	180	3	23	23	NUM
ijassa-348	180	4	3.5	3.5	NUM
ijassa-348	180	5	linear	linear	NOUN
ijassa-348	180	6	discriminant	discriminant	NOUN
ijassa-348	180	7	classifier	classifier	NOUN
ijassa-348	180	8	discriminant	discriminant	ADJ
ijassa-348	180	9	analysis	analysis	NOUN
ijassa-348	180	10	is	be	AUX
ijassa-348	180	11	a	a	DET
ijassa-348	180	12	classification	classification	NOUN
ijassa-348	180	13	problem	problem	NOUN
ijassa-348	180	14	where	where	SCONJ
ijassa-348	180	15	more	more	ADJ
ijassa-348	180	16	than	than	ADP
ijassa-348	180	17	two	two	NUM
ijassa-348	180	18	groups	group	NOUN
ijassa-348	180	19	of	of	ADP
ijassa-348	180	20	populations	population	NOUN
ijassa-348	180	21	are	be	AUX
ijassa-348	180	22	known	know	VERB
ijassa-348	180	23	a	a	DET
ijassa-348	180	24	priori	priori	NOUN
ijassa-348	180	25	and	and	CCONJ
ijassa-348	180	26	one	one	NUM
ijassa-348	180	27	or	or	CCONJ
ijassa-348	180	28	more	more	ADJ
ijassa-348	180	29	samples	sample	NOUN
ijassa-348	180	30	from	from	ADP
ijassa-348	180	31	the	the	DET
ijassa-348	180	32	population	population	NOUN
ijassa-348	180	33	are	be	AUX
ijassa-348	180	34	classified	classify	VERB
ijassa-348	180	35	according	accord	VERB
ijassa-348	180	36	to	to	ADP
ijassa-348	180	37	the	the	DET
ijassa-348	180	38	characteristics	characteristic	NOUN
ijassa-348	180	39	measured	measure	VERB
ijassa-348	180	40	.	.	PUNCT
ijassa-348	181	1	the	the	DET
ijassa-348	181	2	assumption	assumption	NOUN
ijassa-348	181	3	is	be	AUX
ijassa-348	181	4	that	that	SCONJ
ijassa-348	181	5	the	the	DET
ijassa-348	181	6	population	population	NOUN
ijassa-348	181	7	πi	πi	ADV
ijassa-348	181	8	has	have	VERB
ijassa-348	181	9	a	a	DET
ijassa-348	181	10	probability	probability	NOUN
ijassa-348	181	11	density	density	NOUN
ijassa-348	181	12	function	function	NOUN
ijassa-348	181	13	of	of	ADP
ijassa-348	181	14	x	x	PUNCT
ijassa-348	181	15	which	which	PRON
ijassa-348	181	16	has	have	VERB
ijassa-348	181	17	a	a	DET
ijassa-348	181	18	mean	mean	ADJ
ijassa-348	181	19	vector	vector	NOUN
ijassa-348	181	20	ui	ui	NOUN
ijassa-348	181	21	and	and	CCONJ
ijassa-348	181	22	variance	variance	NOUN
ijassa-348	181	23	-	-	PUNCT
ijassa-348	181	24	covariance	covariance	NOUN
ijassa-348	181	25	matrix	matrix	NOUN
ijassa-348	181	26	σ	σ	NOUN
ijassa-348	181	27	(	(	PUNCT
ijassa-348	181	28	similar	similar	ADJ
ijassa-348	181	29	for	for	ADP
ijassa-348	181	30	all	all	DET
ijassa-348	181	31	populations	population	NOUN
ijassa-348	181	32	)	)	PUNCT
ijassa-348	181	33	.	.	PUNCT
ijassa-348	182	1	it	it	PRON
ijassa-348	182	2	is	be	AUX
ijassa-348	182	3	specified	specify	VERB
ijassa-348	182	4	as	as	ADP
ijassa-348	182	5	f	f	PROPN
ijassa-348	182	6	(	(	PUNCT
ijassa-348	182	7	x	x	X
ijassa-348	182	8	|	|	ADV
ijassa-348	182	9	πi	πi	ADP
ijassa-348	182	10	)	)	PUNCT
ijassa-348	182	11	=	=	SYM
ijassa-348	183	1	1	1	NUM
ijassa-348	183	2	|	|	NOUN
ijassa-348	183	3	σ	σ	NUM
ijassa-348	183	4	|1/2(2π)p/2	|1/2(2π)p/2	NOUN
ijassa-348	183	5	exp	exp	NOUN
ijassa-348	183	6	[	[	PUNCT
ijassa-348	183	7	−1	−1	NOUN
ijassa-348	183	8	2	2	NUM
ijassa-348	183	9	(	(	PUNCT
ijassa-348	183	10	x−	x−	PROPN
ijassa-348	183	11	ui	ui	PROPN
ijassa-348	183	12	)	)	PUNCT
ijassa-348	183	13	′	′	NUM
ijassa-348	184	1	σ−1	σ−1	INTJ
ijassa-348	184	2	(	(	PUNCT
ijassa-348	184	3	x−	x−	PROPN
ijassa-348	184	4	ui	ui	PROPN
ijassa-348	184	5	)	)	PUNCT
ijassa-348	184	6	]	]	PUNCT
ijassa-348	184	7	(	(	PUNCT
ijassa-348	184	8	10	10	NUM
ijassa-348	184	9	)	)	PUNCT
ijassa-348	184	10	we	we	PRON
ijassa-348	184	11	classify	classify	VERB
ijassa-348	184	12	to	to	ADP
ijassa-348	184	13	the	the	DET
ijassa-348	184	14	population	population	NOUN
ijassa-348	184	15	for	for	ADP
ijassa-348	184	16	which	which	PRON
ijassa-348	184	17	p	p	X
ijassa-348	184	18	i	i	PRON
ijassa-348	184	19	f(x	f(x	PROPN
ijassa-348	184	20	—	—	PUNCT
ijassa-348	184	21	i	i	PROPN
ijassa-348	184	22	)	)	PUNCT
ijassa-348	184	23	is	be	AUX
ijassa-348	184	24	the	the	DET
ijassa-348	184	25	highest	high	ADJ
ijassa-348	184	26	.	.	PUNCT
ijassa-348	185	1	lda	lda	PROPN
ijassa-348	185	2	is	be	AUX
ijassa-348	185	3	used	use	VERB
ijassa-348	185	4	when	when	SCONJ
ijassa-348	185	5	the	the	DET
ijassa-348	185	6	variance	variance	NOUN
ijassa-348	185	7	-	-	PUNCT
ijassa-348	185	8	covariance	covariance	NOUN
ijassa-348	185	9	matrix	matrix	NOUN
ijassa-348	185	10	is	be	AUX
ijassa-348	185	11	not	not	PART
ijassa-348	185	12	dependent	dependent	ADJ
ijassa-348	185	13	on	on	ADP
ijassa-348	185	14	the	the	DET
ijassa-348	185	15	population	population	NOUN
ijassa-348	185	16	from	from	ADP
ijassa-348	185	17	the	the	DET
ijassa-348	185	18	available	available	ADJ
ijassa-348	185	19	data	datum	NOUN
ijassa-348	185	20	.	.	PUNCT
ijassa-348	186	1	in	in	ADP
ijassa-348	186	2	such	such	ADJ
ijassa-348	186	3	cases	case	NOUN
ijassa-348	186	4	the	the	DET
ijassa-348	186	5	decision	decision	NOUN
ijassa-348	186	6	rule	rule	NOUN
ijassa-348	186	7	is	be	AUX
ijassa-348	186	8	based	base	VERB
ijassa-348	186	9	on	on	ADP
ijassa-348	186	10	the	the	DET
ijassa-348	186	11	linear	linear	ADJ
ijassa-348	186	12	score	score	NOUN
ijassa-348	186	13	function	function	NOUN
ijassa-348	186	14	which	which	PRON
ijassa-348	186	15	is	be	AUX
ijassa-348	186	16	a	a	DET
ijassa-348	186	17	function	function	NOUN
ijassa-348	186	18	of	of	ADP
ijassa-348	186	19	the	the	DET
ijassa-348	186	20	means	mean	NOUN
ijassa-348	186	21	for	for	ADP
ijassa-348	186	22	each	each	PRON
ijassa-348	186	23	of	of	ADP
ijassa-348	186	24	our	our	PRON
ijassa-348	186	25	g	g	PROPN
ijassa-348	186	26	population	population	NOUN
ijassa-348	186	27	ui	ui	NOUN
ijassa-348	187	1	and	and	CCONJ
ijassa-348	187	2	also	also	ADV
ijassa-348	187	3	the	the	DET
ijassa-348	187	4	variance	variance	NOUN
ijassa-348	187	5	ccovariance	ccovariance	NOUN
ijassa-348	187	6	matrix	matrix	NOUN
ijassa-348	187	7	.	.	PUNCT
ijassa-348	188	1	the	the	DET
ijassa-348	188	2	linear	linear	ADJ
ijassa-348	188	3	score	score	NOUN
ijassa-348	188	4	function	function	NOUN
ijassa-348	188	5	is	be	AUX
ijassa-348	188	6	as	as	SCONJ
ijassa-348	188	7	follows	follow	VERB
ijassa-348	188	8	:	:	PUNCT
ijassa-348	189	1	si	si	PROPN
ijassa-348	189	2	l	l	NOUN
ijassa-348	189	3	(	(	PUNCT
ijassa-348	189	4	x	x	X
ijassa-348	189	5	)	)	PUNCT
ijassa-348	189	6	=	=	SYM
ijassa-348	189	7	−1	−1	NOUN
ijassa-348	189	8	2	2	NUM
ijassa-348	189	9	ui	ui	NOUN
ijassa-348	189	10	′	′	NUM
ijassa-348	189	11	σ−1ui	σ−1ui	NOUN
ijassa-348	189	12	′	′	NUM
ijassa-348	190	1	+	+	CCONJ
ijassa-348	190	2	ui	ui	PROPN
ijassa-348	190	3	′	′	NUM
ijassa-348	190	4	σ−1x+	σ−1x+	PROPN
ijassa-348	190	5	logpi	logpi	NOUN
ijassa-348	190	6	=	=	PUNCT
ijassa-348	190	7	d̂i0	d̂i0	NOUN
ijassa-348	190	8	+	+	CCONJ
ijassa-348	190	9	p∑	p∑	X
ijassa-348	191	1	j=1	j=1	ADJ
ijassa-348	191	2	d̂ijxj	d̂ijxj	ADJ
ijassa-348	191	3	+	+	X
ijassa-348	191	4	logpi	logpi	NOUN
ijassa-348	191	5	(	(	PUNCT
ijassa-348	191	6	11	11	NUM
ijassa-348	191	7	)	)	PUNCT
ijassa-348	191	8	where	where	SCONJ
ijassa-348	191	9	di0	di0	NOUN
ijassa-348	191	10	=	=	SYM
ijassa-348	191	11	−1	−1	NOUN
ijassa-348	191	12	2ui	2ui	NOUN
ijassa-348	192	1	′	′	NUM
ijassa-348	192	2	σ−1uiui	σ−1uiui	INTJ
ijassa-348	193	1	′	′	INTJ
ijassa-348	194	1	σ−1	σ−1	INTJ
ijassa-348	194	2	dij	dij	INTJ
ijassa-348	195	1	=	=	SYM
ijassa-348	195	2	jth	jth	PROPN
ijassa-348	195	3	element	element	NOUN
ijassa-348	195	4	of	of	ADP
ijassa-348	195	5	ui	ui	PROPN
ijassa-348	196	1	′	′	NUM
ijassa-348	197	1	σ−1	σ−1	INTJ
ijassa-348	197	2	the	the	DET
ijassa-348	197	3	far	far	ADV
ijassa-348	197	4	left	left	ADJ
ijassa-348	197	5	hand	hand	NOUN
ijassa-348	197	6	expression	expression	NOUN
ijassa-348	197	7	represents	represent	VERB
ijassa-348	197	8	a	a	DET
ijassa-348	197	9	linear	linear	ADJ
ijassa-348	197	10	regression	regression	NOUN
ijassa-348	197	11	with	with	ADP
ijassa-348	197	12	intercept	intercept	NOUN
ijassa-348	197	13	di0	di0	PROPN
ijassa-348	197	14	and	and	CCONJ
ijassa-348	197	15	regression	regression	NOUN
ijassa-348	197	16	coefficients	coefficient	NOUN
ijassa-348	197	17	dij	dij	VERB
ijassa-348	197	18	.	.	PUNCT
ijassa-348	198	1	di	di	INTJ
ijassa-348	198	2	l	l	NOUN
ijassa-348	198	3	(	(	PUNCT
ijassa-348	198	4	x	x	X
ijassa-348	198	5	)	)	PUNCT
ijassa-348	198	6	=	=	SYM
ijassa-348	198	7	−1	−1	NOUN
ijassa-348	198	8	2ui	2ui	NOUN
ijassa-348	198	9	′	′	NUM
ijassa-348	199	1	σ−1ui	σ−1ui	NOUN
ijassa-348	199	2	′	′	NUM
ijassa-348	200	1	+	+	CCONJ
ijassa-348	200	2	ui	ui	NOUN
ijassa-348	201	1	′	′	NOUN
ijassa-348	201	2	σ−1x	σ−1x	NOUN
ijassa-348	201	3	=	=	SYM
ijassa-348	201	4	d̂i0	d̂i0	NOUN
ijassa-348	201	5	+	+	CCONJ
ijassa-348	202	1	∑p	∑p	ADJ
ijassa-348	202	2	j=1	j=1	ADJ
ijassa-348	202	3	d̂ijxj	d̂ijxj	NUM
ijassa-348	202	4	given	give	VERB
ijassa-348	202	5	a	a	DET
ijassa-348	202	6	sample	sample	NOUN
ijassa-348	202	7	unit	unit	NOUN
ijassa-348	202	8	with	with	ADP
ijassa-348	202	9	measurements	measurement	NOUN
ijassa-348	202	10	x1,x2	x1,x2	PROPN
ijassa-348	202	11	·	·	PUNCT
ijassa-348	202	12	xp	xp	INTJ
ijassa-348	202	13	,	,	PUNCT
ijassa-348	202	14	the	the	DET
ijassa-348	202	15	sample	sample	NOUN
ijassa-348	202	16	unit	unit	NOUN
ijassa-348	202	17	is	be	AUX
ijassa-348	202	18	classified	classify	VERB
ijassa-348	202	19	into	into	ADP
ijassa-348	202	20	the	the	DET
ijassa-348	202	21	population	population	NOUN
ijassa-348	202	22	that	that	PRON
ijassa-348	202	23	has	have	VERB
ijassa-348	202	24	the	the	DET
ijassa-348	202	25	highest	high	ADJ
ijassa-348	202	26	linear	linear	ADJ
ijassa-348	202	27	score	score	NOUN
ijassa-348	202	28	.	.	PUNCT
ijassa-348	203	1	this	this	PRON
ijassa-348	203	2	is	be	AUX
ijassa-348	203	3	comparable	comparable	ADJ
ijassa-348	203	4	to	to	ADP
ijassa-348	203	5	the	the	DET
ijassa-348	203	6	population	population	NOUN
ijassa-348	203	7	that	that	PRON
ijassa-348	203	8	has	have	VERB
ijassa-348	203	9	the	the	DET
ijassa-348	203	10	highest	high	ADJ
ijassa-348	203	11	membership	membership	NOUN
ijassa-348	203	12	of	of	ADP
ijassa-348	203	13	posterior	posterior	ADJ
ijassa-348	203	14	probability	probability	NOUN
ijassa-348	203	15	.	.	PUNCT
ijassa-348	204	1	linear	linear	ADJ
ijassa-348	204	2	score	score	NOUN
ijassa-348	204	3	has	have	VERB
ijassa-348	204	4	to	to	PART
ijassa-348	204	5	be	be	AUX
ijassa-348	204	6	calculated	calculate	VERB
ijassa-348	204	7	for	for	ADP
ijassa-348	204	8	each	each	DET
ijassa-348	204	9	class	class	NOUN
ijassa-348	204	10	of	of	ADP
ijassa-348	204	11	population	population	NOUN
ijassa-348	204	12	and	and	CCONJ
ijassa-348	204	13	then	then	ADV
ijassa-348	204	14	the	the	DET
ijassa-348	204	15	assignment	assignment	NOUN
ijassa-348	204	16	of	of	ADP
ijassa-348	204	17	the	the	DET
ijassa-348	204	18	sample	sample	NOUN
ijassa-348	204	19	to	to	ADP
ijassa-348	204	20	the	the	DET
ijassa-348	204	21	population	population	NOUN
ijassa-348	204	22	with	with	ADP
ijassa-348	204	23	highest	high	ADJ
ijassa-348	204	24	score	score	NOUN
ijassa-348	204	25	.	.	PUNCT
ijassa-348	205	1	but	but	CCONJ
ijassa-348	205	2	as	as	SCONJ
ijassa-348	205	3	this	this	DET
ijassa-348	205	4	function	function	NOUN
ijassa-348	205	5	utilizes	utilize	VERB
ijassa-348	205	6	unknown	unknown	ADJ
ijassa-348	205	7	parameters	parameter	NOUN
ijassa-348	205	8	ui	ui	PROPN
ijassa-348	205	9	and	and	CCONJ
ijassa-348	205	10	σ	σ	NOUN
ijassa-348	205	11	these	these	DET
ijassa-348	205	12	parameters	parameter	NOUN
ijassa-348	205	13	have	have	VERB
ijassa-348	205	14	to	to	PART
ijassa-348	205	15	be	be	AUX
ijassa-348	205	16	determined	determine	VERB
ijassa-348	205	17	from	from	ADP
ijassa-348	205	18	the	the	DET
ijassa-348	205	19	data	datum	NOUN
ijassa-348	205	20	.	.	PUNCT
ijassa-348	206	1	hence	hence	ADV
ijassa-348	206	2	discriminant	discriminant	ADJ
ijassa-348	206	3	analysis	analysis	NOUN
ijassa-348	206	4	requires	require	VERB
ijassa-348	206	5	estimation	estimation	NOUN
ijassa-348	206	6	of	of	ADP
ijassa-348	206	7	the	the	DET
ijassa-348	206	8	following	following	ADJ
ijassa-348	206	9	prior	prior	ADJ
ijassa-348	206	10	probabilities	probability	NOUN
ijassa-348	206	11	:	:	PUNCT
ijassa-348	206	12	pi	pi	NOUN
ijassa-348	206	13	=	=	PUNCT
ijassa-348	206	14	pr	pr	X
ijassa-348	206	15	(	(	PUNCT
ijassa-348	206	16	πi	πi	PROPN
ijassa-348	206	17	)	)	PUNCT
ijassa-348	206	18	;	;	PUNCT
ijassa-348	207	1	i	i	NOUN
ijassa-348	207	2	=	=	NOUN
ijassa-348	207	3	1	1	NUM
ijassa-348	207	4	,	,	PUNCT
ijassa-348	207	5	2	2	NUM
ijassa-348	207	6	,	,	PUNCT
ijassa-348	207	7	...	...	PUNCT
ijassa-348	207	8	g	g	NOUN
ijassa-348	207	9	population	population	NOUN
ijassa-348	207	10	means	mean	VERB
ijassa-348	207	11	:	:	PUNCT
ijassa-348	207	12	these	these	PRON
ijassa-348	207	13	can	can	AUX
ijassa-348	207	14	be	be	AUX
ijassa-348	207	15	determined	determine	VERB
ijassa-348	207	16	by	by	ADP
ijassa-348	207	17	sample	sample	NOUN
ijassa-348	207	18	vectors	vector	NOUN
ijassa-348	207	19	.	.	PUNCT
ijassa-348	208	1	ui	ui	NOUN
ijassa-348	209	1	=	=	PUNCT
ijassa-348	209	2	e	e	X
ijassa-348	209	3	(	(	PUNCT
ijassa-348	209	4	x	x	X
ijassa-348	209	5	|	|	ADV
ijassa-348	209	6	πi	πi	ADP
ijassa-348	209	7	)	)	PUNCT
ijassa-348	209	8	;	;	PUNCT
ijassa-348	209	9	i	i	NOUN
ijassa-348	209	10	=	=	NOUN
ijassa-348	209	11	1	1	NUM
ijassa-348	209	12	,	,	PUNCT
ijassa-348	209	13	2	2	NUM
ijassa-348	209	14	,	,	PUNCT
ijassa-348	209	15	...	...	PUNCT
ijassa-348	209	16	g	g	PROPN
ijassa-348	209	17	variance	variance	NOUN
ijassa-348	209	18	-	-	PUNCT
ijassa-348	209	19	covariance	covariance	NOUN
ijassa-348	209	20	matrix	matrix	NOUN
ijassa-348	209	21	:	:	PUNCT
ijassa-348	209	22	this	this	PRON
ijassa-348	209	23	can	can	AUX
ijassa-348	209	24	be	be	AUX
ijassa-348	209	25	determined	determine	VERB
ijassa-348	209	26	using	use	VERB
ijassa-348	209	27	the	the	DET
ijassa-348	209	28	pooled	pooled	ADJ
ijassa-348	209	29	variancecovariance	variancecovariance	NOUN
ijassa-348	209	30	matrix	matrix	NOUN
ijassa-348	209	31	.	.	PUNCT
ijassa-348	210	1	24	24	NUM
ijassa-348	210	2	sumaiya	sumaiya	PROPN
ijassa-348	210	3	thaseen	thaseen	PROPN
ijassa-348	210	4	and	and	CCONJ
ijassa-348	210	5	ch.aswani	ch.aswani	X
ijassa-348	211	1	kumar	kumar	PROPN
ijassa-348	211	2	:	:	PUNCT
ijassa-348	211	3	intrusion	intrusion	NOUN
ijassa-348	211	4	detection	detection	NOUN
ijassa-348	211	5	model	model	NOUN
ijassa-348	211	6	using	use	VERB
ijassa-348	211	7	pca	pca	PROPN
ijassa-348	211	8	and	and	CCONJ
ijassa-348	211	9	...	...	PUNCT
ijassa-348	212	1	σ	σ	NOUN
ijassa-348	212	2	=	=	PUNCT
ijassa-348	212	3	var	var	NOUN
ijassa-348	212	4	(	(	PUNCT
ijassa-348	212	5	x	x	X
ijassa-348	212	6	|	|	ADV
ijassa-348	212	7	πi	πi	ADP
ijassa-348	212	8	)	)	PUNCT
ijassa-348	212	9	;	;	PUNCT
ijassa-348	212	10	i	i	NOUN
ijassa-348	212	11	=	=	NOUN
ijassa-348	212	12	1	1	NUM
ijassa-348	212	13	,	,	PUNCT
ijassa-348	212	14	2	2	NUM
ijassa-348	212	15	,	,	PUNCT
ijassa-348	212	16	...	...	PUNCT
ijassa-348	212	17	g	g	NOUN
ijassa-348	212	18	(	(	PUNCT
ijassa-348	212	19	12	12	NUM
ijassa-348	212	20	)	)	PUNCT
ijassa-348	212	21	generally	generally	ADV
ijassa-348	212	22	these	these	DET
ijassa-348	212	23	parameters	parameter	NOUN
ijassa-348	212	24	are	be	AUX
ijassa-348	212	25	determined	determine	VERB
ijassa-348	212	26	from	from	ADP
ijassa-348	212	27	training	training	NOUN
ijassa-348	212	28	set	set	NOUN
ijassa-348	212	29	for	for	ADP
ijassa-348	212	30	which	which	PRON
ijassa-348	212	31	the	the	DET
ijassa-348	212	32	population	population	NOUN
ijassa-348	212	33	membership	membership	NOUN
ijassa-348	212	34	is	be	AUX
ijassa-348	212	35	already	already	ADV
ijassa-348	212	36	calculated	calculate	VERB
ijassa-348	212	37	.	.	PUNCT
ijassa-348	213	1	conditional	conditional	ADJ
ijassa-348	213	2	density	density	NOUN
ijassa-348	213	3	function	function	NOUN
ijassa-348	213	4	parameters	parameter	NOUN
ijassa-348	213	5	population	population	NOUN
ijassa-348	213	6	means	mean	VERB
ijassa-348	213	7	:	:	PUNCT
ijassa-348	213	8	i	i	PRON
ijassa-348	213	9	can	can	AUX
ijassa-348	213	10	be	be	AUX
ijassa-348	213	11	determined	determine	VERB
ijassa-348	213	12	by	by	ADP
ijassa-348	213	13	replacing	replace	VERB
ijassa-348	213	14	in	in	ADP
ijassa-348	213	15	the	the	DET
ijassa-348	213	16	sample	sample	NOUN
ijassa-348	213	17	means	mean	VERB
ijassa-348	213	18	xi	xi	ADP
ijassa-348	213	19	variance	variance	NOUN
ijassa-348	213	20	-	-	PUNCT
ijassa-348	213	21	covariance	covariance	NOUN
ijassa-348	213	22	matrix	matrix	NOUN
ijassa-348	213	23	:	:	PUNCT
ijassa-348	213	24	let	let	VERB
ijassa-348	213	25	s	s	PRON
ijassa-348	213	26	i	i	PRON
ijassa-348	213	27	represent	represent	VERB
ijassa-348	213	28	the	the	DET
ijassa-348	213	29	sample	sample	NOUN
ijassa-348	213	30	variance	variance	NOUN
ijassa-348	213	31	-	-	PUNCT
ijassa-348	213	32	covariance	covariance	NOUN
ijassa-348	213	33	matrix	matrix	NOUN
ijassa-348	213	34	for	for	ADP
ijassa-348	213	35	the	the	DET
ijassa-348	213	36	population	population	NOUN
ijassa-348	213	37	i.	i.	NOUN
ijassa-348	213	38	thus	thus	ADV
ijassa-348	213	39	the	the	DET
ijassa-348	213	40	variance	variance	NOUN
ijassa-348	213	41	-	-	PUNCT
ijassa-348	213	42	covariance	covariance	NOUN
ijassa-348	213	43	matrix	matrix	NOUN
ijassa-348	213	44	σ	σ	NOUN
ijassa-348	213	45	can	can	AUX
ijassa-348	213	46	be	be	AUX
ijassa-348	213	47	determined	determine	VERB
ijassa-348	213	48	by	by	ADP
ijassa-348	213	49	substituting	substitute	VERB
ijassa-348	213	50	the	the	DET
ijassa-348	213	51	pooled	pooled	ADJ
ijassa-348	213	52	variance	variance	NOUN
ijassa-348	213	53	-	-	NOUN
ijassa-348	213	54	covariance	covariance	NOUN
ijassa-348	213	55	into	into	ADP
ijassa-348	213	56	the	the	DET
ijassa-348	213	57	linear	linear	ADJ
ijassa-348	213	58	score	score	NOUN
ijassa-348	213	59	as	as	SCONJ
ijassa-348	213	60	given	give	VERB
ijassa-348	213	61	below	below	ADV
ijassa-348	213	62	:	:	PUNCT
ijassa-348	213	63	sp	sp	ADP
ijassa-348	213	64	=	=	PUNCT
ijassa-348	213	65	∑g	∑g	ADJ
ijassa-348	213	66	i=1	i=1	PROPN
ijassa-348	214	1	(	(	PUNCT
ijassa-348	214	2	ni	ni	PROPN
ijassa-348	214	3	−	−	PROPN
ijassa-348	215	1	1)si∑g	1)si∑g	INTJ
ijassa-348	215	2	i=1	i=1	PROPN
ijassa-348	216	1	(	(	PUNCT
ijassa-348	216	2	ni	ni	PROPN
ijassa-348	216	3	−	−	PROPN
ijassa-348	216	4	1	1	NUM
ijassa-348	216	5	)	)	PUNCT
ijassa-348	216	6	(	(	PUNCT
ijassa-348	216	7	13	13	NUM
ijassa-348	216	8	)	)	PUNCT
ijassa-348	216	9	to	to	PART
ijassa-348	216	10	obtain	obtain	VERB
ijassa-348	216	11	the	the	DET
ijassa-348	216	12	linear	linear	ADJ
ijassa-348	216	13	score	score	NOUN
ijassa-348	216	14	,	,	PUNCT
ijassa-348	216	15	ˆsil	ˆsil	NOUN
ijassa-348	216	16	(	(	PUNCT
ijassa-348	216	17	x	x	NOUN
ijassa-348	216	18	)	)	PUNCT
ijassa-348	216	19	=	=	SYM
ijassa-348	216	20	−1	−1	NOUN
ijassa-348	217	1	2xisp	2xisp	NUM
ijassa-348	217	2	−1xi	−1xi	PROPN
ijassa-348	217	3	+	+	CCONJ
ijassa-348	217	4	ui	ui	PROPN
ijassa-348	217	5	′	′	NUM
ijassa-348	217	6	σ−1x+	σ−1x+	PROPN
ijassa-348	217	7	logpi	logpi	NOUN
ijassa-348	217	8	=	=	PUNCT
ijassa-348	217	9	d̂i0	d̂i0	NOUN
ijassa-348	217	10	+	+	CCONJ
ijassa-348	218	1	∑p	∑p	ADJ
ijassa-348	218	2	j=1	j=1	ADJ
ijassa-348	218	3	d̂ijxj	d̂ijxj	NOUN
ijassa-348	218	4	+	+	CCONJ
ijassa-348	218	5	logpi	logpi	NOUN
ijassa-348	218	6	where	where	SCONJ
ijassa-348	218	7	,	,	PUNCT
ijassa-348	218	8	d̂i0	d̂i0	NOUN
ijassa-348	218	9	=	=	SYM
ijassa-348	218	10	−1	−1	NOUN
ijassa-348	218	11	2xisp	2xisp	NUM
ijassa-348	218	12	−1xi	−1xi	PROPN
ijassa-348	218	13	and	and	CCONJ
ijassa-348	218	14	dij	dij	PROPN
ijassa-348	218	15	=	=	SYM
ijassa-348	218	16	jth	jth	PROPN
ijassa-348	218	17	element	element	NOUN
ijassa-348	218	18	of	of	ADP
ijassa-348	218	19	xisp	xisp	NUM
ijassa-348	218	20	−1	−1	NOUN
ijassa-348	219	1	this	this	PRON
ijassa-348	219	2	is	be	AUX
ijassa-348	219	3	a	a	DET
ijassa-348	219	4	function	function	NOUN
ijassa-348	219	5	of	of	ADP
ijassa-348	219	6	the	the	DET
ijassa-348	219	7	sample	sample	NOUN
ijassa-348	219	8	mean	mean	NOUN
ijassa-348	219	9	vectors	vector	NOUN
ijassa-348	219	10	,	,	PUNCT
ijassa-348	219	11	the	the	DET
ijassa-348	219	12	variance	variance	NOUN
ijassa-348	219	13	-	-	PUNCT
ijassa-348	219	14	covariance	covariance	NOUN
ijassa-348	219	15	matrix	matrix	NOUN
ijassa-348	219	16	and	and	CCONJ
ijassa-348	219	17	prior	prior	ADJ
ijassa-348	219	18	probabilities	probability	NOUN
ijassa-348	219	19	for	for	ADP
ijassa-348	219	20	different	different	ADJ
ijassa-348	219	21	populations	population	NOUN
ijassa-348	219	22	.	.	PUNCT
ijassa-348	220	1	thus	thus	ADV
ijassa-348	220	2	the	the	DET
ijassa-348	220	3	expression	expression	NOUN
ijassa-348	220	4	looks	look	VERB
ijassa-348	220	5	similar	similar	ADJ
ijassa-348	220	6	to	to	AUX
ijassa-348	220	7	linear	linear	VERB
ijassa-348	220	8	regression	regression	NOUN
ijassa-348	220	9	formula	formula	NOUN
ijassa-348	220	10	with	with	ADP
ijassa-348	220	11	a	a	DET
ijassa-348	220	12	term	term	NOUN
ijassa-348	220	13	for	for	ADP
ijassa-348	220	14	intercept	intercept	NOUN
ijassa-348	220	15	and	and	CCONJ
ijassa-348	220	16	a	a	DET
ijassa-348	220	17	linear	linear	ADJ
ijassa-348	220	18	combination	combination	NOUN
ijassa-348	220	19	of	of	ADP
ijassa-348	220	20	response	response	NOUN
ijassa-348	220	21	variables	variable	NOUN
ijassa-348	220	22	with	with	ADP
ijassa-348	220	23	the	the	DET
ijassa-348	220	24	natural	natural	ADJ
ijassa-348	220	25	log	log	NOUN
ijassa-348	220	26	of	of	ADP
ijassa-348	220	27	the	the	DET
ijassa-348	220	28	probabilities	probability	NOUN
ijassa-348	220	29	.	.	PUNCT
ijassa-348	221	1	thus	thus	ADV
ijassa-348	221	2	the	the	DET
ijassa-348	221	3	decision	decision	NOUN
ijassa-348	221	4	rule	rule	NOUN
ijassa-348	221	5	is	be	AUX
ijassa-348	221	6	to	to	PART
ijassa-348	221	7	classify	classify	VERB
ijassa-348	221	8	the	the	DET
ijassa-348	221	9	sample	sample	NOUN
ijassa-348	221	10	item	item	NOUN
ijassa-348	221	11	into	into	ADP
ijassa-348	221	12	the	the	DET
ijassa-348	221	13	population	population	NOUN
ijassa-348	221	14	that	that	PRON
ijassa-348	221	15	has	have	VERB
ijassa-348	221	16	the	the	DET
ijassa-348	221	17	highest	high	ADJ
ijassa-348	221	18	calculated	calculate	VERB
ijassa-348	221	19	linear	linear	PROPN
ijassa-348	221	20	score	score	NOUN
ijassa-348	221	21	.	.	PUNCT
ijassa-348	222	1	the	the	DET
ijassa-348	222	2	steps	step	NOUN
ijassa-348	222	3	to	to	PART
ijassa-348	222	4	identify	identify	VERB
ijassa-348	222	5	the	the	DET
ijassa-348	222	6	class	class	NOUN
ijassa-348	222	7	label	label	NOUN
ijassa-348	222	8	is	be	AUX
ijassa-348	222	9	as	as	SCONJ
ijassa-348	222	10	follows	follow	VERB
ijassa-348	222	11	step	step	NOUN
ijassa-348	222	12	1	1	NUM
ijassa-348	222	13	:	:	PUNCT
ijassa-348	222	14	delete	delete	VERB
ijassa-348	222	15	one	one	NUM
ijassa-348	222	16	observation	observation	NOUN
ijassa-348	222	17	from	from	ADP
ijassa-348	222	18	the	the	DET
ijassa-348	222	19	sample	sample	NOUN
ijassa-348	222	20	.	.	PUNCT
ijassa-348	223	1	step	step	NOUN
ijassa-348	223	2	2	2	NUM
ijassa-348	223	3	:	:	PUNCT
ijassa-348	223	4	compute	compute	VERB
ijassa-348	223	5	the	the	DET
ijassa-348	223	6	discriminant	discriminant	NOUN
ijassa-348	223	7	function	function	VERB
ijassa-348	223	8	using	use	VERB
ijassa-348	223	9	the	the	DET
ijassa-348	223	10	remaining	remain	VERB
ijassa-348	223	11	observations	observation	NOUN
ijassa-348	223	12	.	.	PUNCT
ijassa-348	224	1	step	step	NOUN
ijassa-348	224	2	3	3	NUM
ijassa-348	224	3	:	:	PUNCT
ijassa-348	224	4	calculate	calculate	VERB
ijassa-348	224	5	the	the	DET
ijassa-348	224	6	discriminant	discriminant	NOUN
ijassa-348	224	7	function	function	VERB
ijassa-348	224	8	from	from	ADP
ijassa-348	224	9	step	step	NOUN
ijassa-348	224	10	2	2	NUM
ijassa-348	224	11	to	to	PART
ijassa-348	224	12	identify	identify	VERB
ijassa-348	224	13	the	the	DET
ijassa-348	224	14	class	class	NOUN
ijassa-348	224	15	label	label	NOUN
ijassa-348	224	16	of	of	ADP
ijassa-348	224	17	the	the	DET
ijassa-348	224	18	observation	observation	NOUN
ijassa-348	224	19	removed	remove	VERB
ijassa-348	224	20	from	from	ADP
ijassa-348	224	21	sample	sample	NOUN
ijassa-348	224	22	in	in	ADP
ijassa-348	224	23	step	step	NOUN
ijassa-348	224	24	1	1	NUM
ijassa-348	224	25	.	.	PUNCT
ijassa-348	225	1	steps	step	NOUN
ijassa-348	225	2	1	1	NUM
ijassa-348	225	3	-	-	SYM
ijassa-348	225	4	3	3	NUM
ijassa-348	225	5	are	be	AUX
ijassa-348	225	6	repeated	repeat	VERB
ijassa-348	225	7	for	for	ADP
ijassa-348	225	8	all	all	DET
ijassa-348	225	9	the	the	DET
ijassa-348	225	10	samples	sample	NOUN
ijassa-348	225	11	.	.	PUNCT
ijassa-348	226	1	calculate	calculate	VERB
ijassa-348	226	2	the	the	DET
ijassa-348	226	3	misclassified	misclassifie	VERB
ijassa-348	226	4	observations	observation	NOUN
ijassa-348	226	5	.	.	PUNCT
ijassa-348	227	1	3.6	3.6	NUM
ijassa-348	227	2	quadratic	quadratic	ADJ
ijassa-348	227	3	discriminant	discriminant	NOUN
ijassa-348	227	4	analysis	analysis	NOUN
ijassa-348	227	5	qda	qda	NOUN
ijassa-348	227	6	is	be	AUX
ijassa-348	227	7	very	very	ADV
ijassa-348	227	8	similar	similar	ADJ
ijassa-348	227	9	to	to	ADP
ijassa-348	227	10	lda	lda	VERB
ijassa-348	227	11	where	where	SCONJ
ijassa-348	227	12	in	in	ADP
ijassa-348	227	13	the	the	DET
ijassa-348	227	14	assumption	assumption	NOUN
ijassa-348	227	15	is	be	AUX
ijassa-348	227	16	that	that	SCONJ
ijassa-348	227	17	the	the	DET
ijassa-348	227	18	each	each	DET
ijassa-348	227	19	class	class	NOUN
ijassa-348	227	20	measurements	measurement	NOUN
ijassa-348	227	21	are	be	AUX
ijassa-348	227	22	distributed	distribute	VERB
ijassa-348	227	23	normally	normally	ADV
ijassa-348	227	24	.	.	PUNCT
ijassa-348	228	1	but	but	CCONJ
ijassa-348	228	2	in	in	ADP
ijassa-348	228	3	qda	qda	NOUN
ijassa-348	228	4	there	there	PRON
ijassa-348	228	5	is	be	VERB
ijassa-348	228	6	no	no	DET
ijassa-348	228	7	such	such	ADJ
ijassa-348	228	8	assumption	assumption	NOUN
ijassa-348	228	9	that	that	SCONJ
ijassa-348	228	10	the	the	DET
ijassa-348	228	11	covariance	covariance	NOUN
ijassa-348	228	12	of	of	ADP
ijassa-348	228	13	each	each	PRON
ijassa-348	228	14	of	of	ADP
ijassa-348	228	15	the	the	DET
ijassa-348	228	16	class	class	NOUN
ijassa-348	228	17	is	be	AUX
ijassa-348	228	18	identical	identical	ADJ
ijassa-348	228	19	.	.	PUNCT
ijassa-348	229	1	if	if	SCONJ
ijassa-348	229	2	the	the	DET
ijassa-348	229	3	normality	normality	NOUN
ijassa-348	229	4	assumption	assumption	NOUN
ijassa-348	229	5	holds	hold	VERB
ijassa-348	229	6	true	true	ADJ
ijassa-348	229	7	,	,	PUNCT
ijassa-348	229	8	then	then	ADV
ijassa-348	229	9	the	the	DET
ijassa-348	229	10	best	good	ADJ
ijassa-348	229	11	possible	possible	ADJ
ijassa-348	229	12	test	test	NOUN
ijassa-348	229	13	for	for	ADP
ijassa-348	229	14	a	a	DET
ijassa-348	229	15	hypothesis	hypothesis	NOUN
ijassa-348	229	16	that	that	PRON
ijassa-348	229	17	the	the	DET
ijassa-348	229	18	given	give	VERB
ijassa-348	229	19	measurement	measurement	NOUN
ijassa-348	229	20	from	from	ADP
ijassa-348	229	21	a	a	DET
ijassa-348	229	22	given	give	VERB
ijassa-348	229	23	class	class	NOUN
ijassa-348	229	24	is	be	AUX
ijassa-348	229	25	named	name	VERB
ijassa-348	229	26	as	as	ADP
ijassa-348	229	27	the	the	DET
ijassa-348	229	28	likelihood	likelihood	NOUN
ijassa-348	229	29	test	test	NOUN
ijassa-348	229	30	.	.	PUNCT
ijassa-348	230	1	assume	assume	VERB
ijassa-348	230	2	that	that	SCONJ
ijassa-348	230	3	there	there	PRON
ijassa-348	230	4	are	be	VERB
ijassa-348	230	5	only	only	ADV
ijassa-348	230	6	2	2	NUM
ijassa-348	230	7	groups	group	NOUN
ijassa-348	230	8	(	(	PUNCT
ijassa-348	230	9	yϵ{0	yϵ{0	ADV
ijassa-348	230	10	,	,	PUNCT
ijassa-348	230	11	1	1	NUM
ijassa-348	230	12	}	}	PUNCT
ijassa-348	230	13	)	)	PUNCT
ijassa-348	230	14	and	and	CCONJ
ijassa-348	230	15	the	the	DET
ijassa-348	230	16	means	mean	NOUN
ijassa-348	230	17	of	of	ADP
ijassa-348	230	18	each	each	DET
ijassa-348	230	19	class	class	NOUN
ijassa-348	230	20	are	be	AUX
ijassa-348	230	21	specified	specify	VERB
ijassa-348	230	22	as	as	ADP
ijassa-348	230	23	uy	uy	PROPN
ijassa-348	230	24	=	=	NOUN
ijassa-348	230	25	0	0	PROPN
ijassa-348	230	26	,	,	PUNCT
ijassa-348	230	27	uy	uy	NOUN
ijassa-348	230	28	=	=	SYM
ijassa-348	230	29	1	1	NUM
ijassa-348	231	1	and	and	CCONJ
ijassa-348	231	2	the	the	DET
ijassa-348	231	3	covariances	covariance	NOUN
ijassa-348	231	4	are	be	AUX
ijassa-348	231	5	defined	define	VERB
ijassa-348	231	6	as	as	ADP
ijassa-348	231	7	σy=0	σy=0	NOUN
ijassa-348	231	8	and	and	CCONJ
ijassa-348	231	9	σy=1	σy=1	NOUN
ijassa-348	231	10	.	.	PUNCT
ijassa-348	232	1	then	then	ADV
ijassa-348	232	2	the	the	DET
ijassa-348	232	3	likelihood	likelihood	NOUN
ijassa-348	232	4	ratio	ratio	NOUN
ijassa-348	232	5	will	will	AUX
ijassa-348	232	6	be	be	AUX
ijassa-348	232	7	specified	specify	VERB
ijassa-348	232	8	as	as	ADP
ijassa-348	232	9	,	,	PUNCT
ijassa-348	232	10	advances	advance	NOUN
ijassa-348	232	11	in	in	ADP
ijassa-348	232	12	systems	system	NOUN
ijassa-348	232	13	science	science	NOUN
ijassa-348	232	14	and	and	CCONJ
ijassa-348	232	15	application	application	NOUN
ijassa-348	232	16	(	(	PUNCT
ijassa-348	232	17	2016	2016	NUM
ijassa-348	232	18	)	)	PUNCT
ijassa-348	233	1	vol.16	vol.16	PROPN
ijassa-348	233	2	no.2	no.2	PROPN
ijassa-348	233	3	25	25	NUM
ijassa-348	233	4	likelihoodratio	likelihoodratio	NOUN
ijassa-348	233	5	=	=	SYM
ijassa-348	233	6	exp	exp	NOUN
ijassa-348	233	7	(	(	PUNCT
ijassa-348	233	8	−1	−1	NOUN
ijassa-348	233	9	2	2	NUM
ijassa-348	233	10	(	(	PUNCT
ijassa-348	233	11	x−	x−	PROPN
ijassa-348	233	12	uy=1	uy=1	PROPN
ijassa-348	233	13	)	)	PUNCT
ijassa-348	233	14	t	t	PROPN
ijassa-348	233	15	∑−1	∑−1	NOUN
ijassa-348	233	16	y=1	y=1	PROPN
ijassa-348	233	17	(	(	PUNCT
ijassa-348	233	18	x−	x−	PROPN
ijassa-348	233	19	uy=1	uy=1	PROPN
ijassa-348	233	20	)	)	PUNCT
ijassa-348	233	21	)	)	PUNCT
ijassa-348	234	1	√	√	ADP
ijassa-348	234	2	2π	2π	NOUN
ijassa-348	235	1	|	|	ADV
ijassa-348	235	2	σy=1	σy=1	INTJ
ijassa-348	235	3	|	|	NOUN
ijassa-348	235	4	−1	−1	NOUN
ijassa-348	235	5	exp	exp	NOUN
ijassa-348	235	6	(	(	PUNCT
ijassa-348	235	7	−1	−1	NOUN
ijassa-348	235	8	2	2	NUM
ijassa-348	235	9	(	(	PUNCT
ijassa-348	235	10	x−	x−	NOUN
ijassa-348	235	11	uy=0	uy=0	NOUN
ijassa-348	235	12	)	)	PUNCT
ijassa-348	235	13	)	)	PUNCT
ijassa-348	236	1	t	t	NOUN
ijassa-348	236	2	∑−1	∑−1	VERB
ijassa-348	236	3	y=0	y=0	PRON
ijassa-348	236	4	√	√	PROPN
ijassa-348	236	5	2π	2π	NOUN
ijassa-348	236	6	|	|	ADV
ijassa-348	236	7	σy=0	σy=0	ADV
ijassa-348	236	8	∥	∥	X
ijassa-348	236	9	−1	−1	NOUN
ijassa-348	236	10	<	<	X
ijassa-348	236	11	t	t	PROPN
ijassa-348	236	12	(	(	PUNCT
ijassa-348	236	13	14	14	NUM
ijassa-348	236	14	)	)	PUNCT
ijassa-348	236	15	for	for	ADP
ijassa-348	236	16	a	a	DET
ijassa-348	236	17	specific	specific	ADJ
ijassa-348	236	18	threshold	threshold	NOUN
ijassa-348	236	19	’	'	PUNCT
ijassa-348	236	20	t	t	PROPN
ijassa-348	236	21	’	'	PUNCT
ijassa-348	236	22	.	.	PUNCT
ijassa-348	237	1	the	the	DET
ijassa-348	237	2	sample	sample	NOUN
ijassa-348	237	3	estimates	estimate	NOUN
ijassa-348	237	4	of	of	ADP
ijassa-348	237	5	the	the	DET
ijassa-348	237	6	mean	mean	ADJ
ijassa-348	237	7	vector	vector	NOUN
ijassa-348	237	8	and	and	CCONJ
ijassa-348	237	9	variance	variance	NOUN
ijassa-348	237	10	-	-	PUNCT
ijassa-348	237	11	covariance	covariance	NOUN
ijassa-348	237	12	matrices	matrix	NOUN
ijassa-348	237	13	will	will	AUX
ijassa-348	237	14	substitute	substitute	VERB
ijassa-348	237	15	the	the	DET
ijassa-348	237	16	population	population	NOUN
ijassa-348	237	17	quantities	quantity	NOUN
ijassa-348	237	18	in	in	ADP
ijassa-348	237	19	the	the	DET
ijassa-348	237	20	formula	formula	NOUN
ijassa-348	237	21	.	.	PUNCT
ijassa-348	238	1	3.7	3.7	NUM
ijassa-348	238	2	ensemble	ensemble	ADJ
ijassa-348	238	3	approach	approach	NOUN
ijassa-348	238	4	using	use	VERB
ijassa-348	238	5	wma	wma	NOUN
ijassa-348	238	6	the	the	DET
ijassa-348	238	7	basic	basic	ADJ
ijassa-348	238	8	idea	idea	NOUN
ijassa-348	238	9	of	of	ADP
ijassa-348	238	10	majority	majority	NOUN
ijassa-348	238	11	voting	voting	NOUN
ijassa-348	238	12	is	be	AUX
ijassa-348	238	13	that	that	SCONJ
ijassa-348	238	14	the	the	DET
ijassa-348	238	15	votes	vote	NOUN
ijassa-348	238	16	are	be	AUX
ijassa-348	238	17	initialized	initialize	VERB
ijassa-348	238	18	to	to	ADP
ijassa-348	238	19	each	each	DET
ijassa-348	238	20	managers	manager	NOUN
ijassa-348	238	21	opinion	opinion	NOUN
ijassa-348	238	22	.	.	PUNCT
ijassa-348	239	1	the	the	DET
ijassa-348	239	2	opinion	opinion	NOUN
ijassa-348	239	3	with	with	ADP
ijassa-348	239	4	the	the	DET
ijassa-348	239	5	highest	high	ADJ
ijassa-348	239	6	votes	vote	NOUN
ijassa-348	239	7	is	be	AUX
ijassa-348	239	8	selected	select	VERB
ijassa-348	239	9	as	as	ADP
ijassa-348	239	10	the	the	DET
ijassa-348	239	11	final	final	ADJ
ijassa-348	239	12	result	result	NOUN
ijassa-348	239	13	.	.	PUNCT
ijassa-348	240	1	littlestone	littlestone	NOUN
ijassa-348	240	2	and	and	CCONJ
ijassa-348	240	3	warmuth	warmuth	NOUN
ijassa-348	241	1	[	[	X
ijassa-348	241	2	37	37	NUM
ijassa-348	241	3	]	]	PUNCT
ijassa-348	241	4	have	have	AUX
ijassa-348	241	5	specified	specify	VERB
ijassa-348	241	6	that	that	SCONJ
ijassa-348	241	7	the	the	DET
ijassa-348	241	8	number	number	NOUN
ijassa-348	241	9	of	of	ADP
ijassa-348	241	10	errors	error	NOUN
ijassa-348	241	11	can	can	AUX
ijassa-348	241	12	be	be	AUX
ijassa-348	241	13	reduced	reduce	VERB
ijassa-348	241	14	in	in	ADP
ijassa-348	241	15	an	an	DET
ijassa-348	241	16	ensemble	ensemble	ADJ
ijassa-348	241	17	model	model	NOUN
ijassa-348	241	18	by	by	ADP
ijassa-348	241	19	introducing	introduce	VERB
ijassa-348	241	20	weights	weight	NOUN
ijassa-348	241	21	to	to	ADP
ijassa-348	241	22	the	the	DET
ijassa-348	241	23	majority	majority	NOUN
ijassa-348	241	24	voting	voting	NOUN
ijassa-348	241	25	technique	technique	NOUN
ijassa-348	241	26	.	.	PUNCT
ijassa-348	242	1	we	we	PRON
ijassa-348	242	2	utilize	utilize	VERB
ijassa-348	242	3	voting	vote	VERB
ijassa-348	242	4	approach	approach	NOUN
ijassa-348	242	5	in	in	ADP
ijassa-348	242	6	this	this	DET
ijassa-348	242	7	paper	paper	NOUN
ijassa-348	242	8	as	as	SCONJ
ijassa-348	242	9	given	give	VERB
ijassa-348	242	10	in	in	ADP
ijassa-348	242	11	[	[	X
ijassa-348	242	12	37	37	NUM
ijassa-348	242	13	]	]	PUNCT
ijassa-348	242	14	.	.	PUNCT
ijassa-348	243	1	each	each	DET
ijassa-348	243	2	manager	manager	NOUN
ijassa-348	243	3	is	be	AUX
ijassa-348	243	4	initialized	initialize	VERB
ijassa-348	243	5	with	with	ADP
ijassa-348	243	6	a	a	DET
ijassa-348	243	7	weight	weight	NOUN
ijassa-348	243	8	obtained	obtain	VERB
ijassa-348	243	9	from	from	ADP
ijassa-348	243	10	managers	manager	NOUN
ijassa-348	243	11	accuracy	accuracy	NOUN
ijassa-348	243	12	in	in	ADP
ijassa-348	243	13	classifying	classify	VERB
ijassa-348	243	14	the	the	DET
ijassa-348	243	15	sample	sample	NOUN
ijassa-348	243	16	.	.	PUNCT
ijassa-348	244	1	as	as	SCONJ
ijassa-348	244	2	each	each	DET
ijassa-348	244	3	manager	manager	NOUN
ijassa-348	244	4	based	base	VERB
ijassa-348	244	5	on	on	ADP
ijassa-348	244	6	the	the	DET
ijassa-348	244	7	dataset	dataset	NOUN
ijassa-348	244	8	contains	contain	VERB
ijassa-348	244	9	different	different	ADJ
ijassa-348	244	10	number	number	NOUN
ijassa-348	244	11	of	of	ADP
ijassa-348	244	12	binary	binary	PROPN
ijassa-348	244	13	classifiers	classifier	NOUN
ijassa-348	244	14	bi	bi	PROPN
ijassa-348	244	15	,	,	PUNCT
ijassa-348	244	16	we	we	PRON
ijassa-348	244	17	need	need	VERB
ijassa-348	244	18	to	to	PART
ijassa-348	244	19	consider	consider	VERB
ijassa-348	244	20	each	each	DET
ijassa-348	244	21	manager	manager	NOUN
ijassa-348	244	22	’s	’s	PART
ijassa-348	244	23	opinion	opinion	NOUN
ijassa-348	244	24	for	for	ADP
ijassa-348	244	25	every	every	DET
ijassa-348	244	26	class	class	NOUN
ijassa-348	244	27	i	i	PRON
ijassa-348	244	28	separately	separately	ADV
ijassa-348	244	29	.	.	PUNCT
ijassa-348	245	1	thus	thus	ADV
ijassa-348	245	2	we	we	PRON
ijassa-348	245	3	can	can	AUX
ijassa-348	245	4	divide	divide	VERB
ijassa-348	245	5	the	the	DET
ijassa-348	245	6	manager	manager	NOUN
ijassa-348	245	7	’s	’s	PART
ijassa-348	245	8	opinion	opinion	NOUN
ijassa-348	245	9	into	into	ADP
ijassa-348	245	10	two	two	NUM
ijassa-348	245	11	categories	category	NOUN
ijassa-348	245	12	:	:	PUNCT
ijassa-348	245	13	•	•	NUM
ijassa-348	245	14	managers	manager	NOUN
ijassa-348	245	15	which	which	PRON
ijassa-348	245	16	classify	classify	VERB
ijassa-348	245	17	given	give	VERB
ijassa-348	245	18	sample	sample	NOUN
ijassa-348	245	19	as	as	ADP
ijassa-348	245	20	an	an	DET
ijassa-348	245	21	object	object	NOUN
ijassa-348	245	22	of	of	ADP
ijassa-348	245	23	class	class	NOUN
ijassa-348	246	1	i	i	PRON
ijassa-348	246	2	(	(	PUNCT
ijassa-348	246	3	output	output	NOUN
ijassa-348	246	4	value	value	NOUN
ijassa-348	246	5	1	1	NUM
ijassa-348	246	6	)	)	PUNCT
ijassa-348	246	7	•	•	NUM
ijassa-348	246	8	managers	manager	NOUN
ijassa-348	246	9	which	which	PRON
ijassa-348	246	10	assert	assert	VERB
ijassa-348	246	11	that	that	SCONJ
ijassa-348	246	12	the	the	DET
ijassa-348	246	13	given	give	VERB
ijassa-348	246	14	observation	observation	NOUN
ijassa-348	246	15	fits	fit	VERB
ijassa-348	246	16	to	to	ADP
ijassa-348	246	17	some	some	DET
ijassa-348	246	18	other	other	ADJ
ijassa-348	246	19	class	class	NOUN
ijassa-348	246	20	than	than	ADP
ijassa-348	246	21	i.	i.	NOUN
ijassa-348	246	22	(	(	PUNCT
ijassa-348	246	23	output	output	NOUN
ijassa-348	246	24	value	value	NOUN
ijassa-348	246	25	0	0	NUM
ijassa-348	246	26	)	)	PUNCT
ijassa-348	246	27	.	.	PUNCT
ijassa-348	247	1	the	the	DET
ijassa-348	247	2	voting	voting	NOUN
ijassa-348	247	3	approach	approach	NOUN
ijassa-348	247	4	is	be	AUX
ijassa-348	247	5	repeated	repeat	VERB
ijassa-348	247	6	for	for	ADP
ijassa-348	247	7	each	each	DET
ijassa-348	247	8	sample	sample	NOUN
ijassa-348	247	9	x	x	PUNCT
ijassa-348	247	10	and	and	CCONJ
ijassa-348	247	11	for	for	ADP
ijassa-348	247	12	each	each	DET
ijassa-348	247	13	binary	binary	ADJ
ijassa-348	247	14	classifier	classifier	NOUN
ijassa-348	247	15	inside	inside	ADP
ijassa-348	247	16	the	the	DET
ijassa-348	247	17	manager	manager	NOUN
ijassa-348	247	18	.	.	PUNCT
ijassa-348	248	1	this	this	DET
ijassa-348	248	2	results	result	VERB
ijassa-348	248	3	in	in	ADP
ijassa-348	248	4	an	an	DET
ijassa-348	248	5	ensemble	ensemble	ADJ
ijassa-348	248	6	manager	manager	NOUN
ijassa-348	248	7	one	one	NUM
ijassa-348	248	8	for	for	ADP
ijassa-348	248	9	each	each	DET
ijassa-348	248	10	class	class	NOUN
ijassa-348	248	11	.	.	PUNCT
ijassa-348	249	1	we	we	PRON
ijassa-348	249	2	define	define	VERB
ijassa-348	249	3	a	a	DET
ijassa-348	249	4	set	set	NOUN
ijassa-348	249	5	of	of	ADP
ijassa-348	249	6	weight	weight	NOUN
ijassa-348	249	7	coefficients	coefficient	NOUN
ijassa-348	249	8	w	w	ADP
ijassa-348	249	9	as	as	ADP
ijassa-348	249	10	a	a	DET
ijassa-348	249	11	15	15	NUM
ijassa-348	249	12	element	element	NOUN
ijassa-348	249	13	vector	vector	NOUN
ijassa-348	249	14	for	for	ADP
ijassa-348	249	15	nsl	nsl	PROPN
ijassa-348	249	16	-	-	PUNCT
ijassa-348	249	17	kdd	kdd	NOUN
ijassa-348	249	18	dataset	dataset	NOUN
ijassa-348	249	19	where	where	SCONJ
ijassa-348	249	20	each	each	DET
ijassa-348	249	21	element	element	NOUN
ijassa-348	249	22	j	j	PROPN
ijassa-348	249	23	represents	represent	VERB
ijassa-348	249	24	weight	weight	NOUN
ijassa-348	249	25	for	for	ADP
ijassa-348	249	26	jth	jth	PROPN
ijassa-348	249	27	manager	manager	NOUN
ijassa-348	249	28	in	in	ADP
ijassa-348	249	29	ensemble	ensemble	ADJ
ijassa-348	249	30	ie	ie	NOUN
ijassa-348	249	31	.	.	PUNCT
ijassa-348	250	1	w	w	NOUN
ijassa-348	250	2	=	=	SYM
ijassa-348	250	3	(	(	PUNCT
ijassa-348	250	4	w1	w1	NOUN
ijassa-348	250	5	,	,	PUNCT
ijassa-348	250	6	w2	w2	NOUN
ijassa-348	250	7	...	...	PUNCT
ijassa-348	250	8	w15	w15	PROPN
ijassa-348	250	9	)	)	PUNCT
ijassa-348	250	10	and	and	CCONJ
ijassa-348	250	11	similarly	similarly	ADV
ijassa-348	250	12	w	w	NOUN
ijassa-348	250	13	as	as	ADP
ijassa-348	250	14	a	a	DET
ijassa-348	250	15	27	27	NUM
ijassa-348	250	16	element	element	NOUN
ijassa-348	250	17	vector	vector	NOUN
ijassa-348	250	18	for	for	ADP
ijassa-348	250	19	unsw	unsw	PROPN
ijassa-348	250	20	-	-	PUNCT
ijassa-348	250	21	nb	nb	NOUN
ijassa-348	250	22	dataset	dataset	NOUN
ijassa-348	250	23	.	.	PUNCT
ijassa-348	251	1	ie	ie	X
ijassa-348	251	2	.	.	PUNCT
ijassa-348	251	3	w	w	NOUN
ijassa-348	251	4	=	=	SYM
ijassa-348	251	5	(	(	PUNCT
ijassa-348	251	6	w1	w1	NOUN
ijassa-348	251	7	,	,	PUNCT
ijassa-348	251	8	w2	w2	NOUN
ijassa-348	251	9	...	...	PUNCT
ijassa-348	251	10	w27	w27	VERB
ijassa-348	251	11	)	)	PUNCT
ijassa-348	251	12	.	.	PUNCT
ijassa-348	252	1	to	to	PART
ijassa-348	252	2	obtain	obtain	VERB
ijassa-348	252	3	the	the	DET
ijassa-348	252	4	final	final	ADJ
ijassa-348	252	5	decision	decision	NOUN
ijassa-348	252	6	function	function	NOUN
ijassa-348	252	7	,	,	PUNCT
ijassa-348	252	8	we	we	PRON
ijassa-348	252	9	consider	consider	VERB
ijassa-348	252	10	the	the	DET
ijassa-348	252	11	weights	weight	NOUN
ijassa-348	252	12	used	use	VERB
ijassa-348	252	13	in	in	ADP
ijassa-348	252	14	the	the	DET
ijassa-348	252	15	voting	voting	NOUN
ijassa-348	252	16	approach	approach	NOUN
ijassa-348	252	17	.	.	PUNCT
ijassa-348	253	1	for	for	ADP
ijassa-348	253	2	every	every	DET
ijassa-348	253	3	single	single	ADJ
ijassa-348	253	4	observation	observation	NOUN
ijassa-348	253	5	x	x	NOUN
ijassa-348	253	6	,	,	PUNCT
ijassa-348	253	7	we	we	PRON
ijassa-348	253	8	obtain	obtain	VERB
ijassa-348	253	9	fifteen	fifteen	NUM
ijassa-348	253	10	output	output	NOUN
ijassa-348	253	11	values	value	NOUN
ijassa-348	253	12	(	(	PUNCT
ijassa-348	253	13	y1	y1	INTJ
ijassa-348	253	14	,	,	PUNCT
ijassa-348	253	15	y2	y2	PROPN
ijassa-348	253	16	...	...	PUNCT
ijassa-348	253	17	y15	y15	NOUN
ijassa-348	253	18	)	)	PUNCT
ijassa-348	253	19	for	for	ADP
ijassa-348	253	20	nsl	nsl	PROPN
ijassa-348	253	21	-	-	PUNCT
ijassa-348	253	22	kdd	kdd	NOUN
ijassa-348	253	23	dataset	dataset	NOUN
ijassa-348	253	24	and	and	CCONJ
ijassa-348	253	25	twenty	twenty	NUM
ijassa-348	253	26	seven	seven	NUM
ijassa-348	253	27	output	output	NOUN
ijassa-348	253	28	values	value	NOUN
ijassa-348	253	29	(	(	PUNCT
ijassa-348	253	30	y1	y1	INTJ
ijassa-348	253	31	,	,	PUNCT
ijassa-348	253	32	y2	y2	PROPN
ijassa-348	253	33	...	...	PUNCT
ijassa-348	253	34	y27	y27	PROPN
ijassa-348	253	35	)	)	PUNCT
ijassa-348	253	36	for	for	ADP
ijassa-348	253	37	unsw	unsw	PROPN
ijassa-348	253	38	-	-	PUNCT
ijassa-348	253	39	nb	nb	NOUN
ijassa-348	253	40	dataset	dataset	NOUN
ijassa-348	253	41	,	,	PUNCT
ijassa-348	253	42	one	one	NUM
ijassa-348	253	43	output	output	NOUN
ijassa-348	253	44	value	value	NOUN
ijassa-348	253	45	per	per	ADP
ijassa-348	253	46	manager	manager	NOUN
ijassa-348	253	47	.	.	PUNCT
ijassa-348	254	1	each	each	DET
ijassa-348	254	2	value	value	NOUN
ijassa-348	254	3	can	can	AUX
ijassa-348	254	4	be	be	AUX
ijassa-348	254	5	a	a	DET
ijassa-348	254	6	positive	positive	ADJ
ijassa-348	254	7	or	or	CCONJ
ijassa-348	254	8	negative	negative	ADJ
ijassa-348	254	9	,	,	PUNCT
ijassa-348	254	10	i.e	i.e	PROPN
ijassa-348	254	11	yj	yj	NOUN
ijassa-348	254	12	=	=	PUNCT
ijassa-348	254	13	{	{	PUNCT
ijassa-348	254	14	−1	−1	NOUN
ijassa-348	254	15	,	,	PUNCT
ijassa-348	254	16	1	1	NUM
ijassa-348	254	17	}	}	PUNCT
ijassa-348	254	18	where	where	SCONJ
ijassa-348	254	19	value	value	NOUN
ijassa-348	254	20	1	1	NUM
ijassa-348	254	21	correspond	correspond	NOUN
ijassa-348	254	22	to	to	ADP
ijassa-348	254	23	managers	manager	NOUN
ijassa-348	254	24	output	output	NOUN
ijassa-348	254	25	1	1	NUM
ijassa-348	254	26	and	and	CCONJ
ijassa-348	254	27	negative	negative	ADJ
ijassa-348	254	28	value	value	NOUN
ijassa-348	254	29	-1	-1	PUNCT
ijassa-348	254	30	represents	represent	VERB
ijassa-348	254	31	managers	manager	NOUN
ijassa-348	254	32	output	output	NOUN
ijassa-348	254	33	0	0	NUM
ijassa-348	254	34	.	.	PUNCT
ijassa-348	255	1	the	the	DET
ijassa-348	255	2	final	final	ADJ
ijassa-348	255	3	decision	decision	NOUN
ijassa-348	255	4	is	be	AUX
ijassa-348	255	5	evaluated	evaluate	VERB
ijassa-348	255	6	by	by	ADP
ijassa-348	255	7	the	the	DET
ijassa-348	255	8	equation	equation	NOUN
ijassa-348	255	9	given	give	VERB
ijassa-348	255	10	below	below	ADP
ijassa-348	255	11	y	y	PROPN
ijassa-348	255	12	=	=	SYM
ijassa-348	255	13	sgn	sgn	PROPN
ijassa-348	255	14	(	(	PUNCT
ijassa-348	255	15	n∑	n∑	NOUN
ijassa-348	255	16	i=1	i=1	PROPN
ijassa-348	255	17	wiyi	wiyi	PROPN
ijassa-348	255	18	)	)	PUNCT
ijassa-348	255	19	(	(	PUNCT
ijassa-348	255	20	15	15	NUM
ijassa-348	255	21	)	)	PUNCT
ijassa-348	255	22	26	26	NUM
ijassa-348	255	23	sumaiya	sumaiya	NOUN
ijassa-348	255	24	thaseen	thaseen	NOUN
ijassa-348	255	25	and	and	CCONJ
ijassa-348	255	26	ch.aswani	ch.aswani	X
ijassa-348	256	1	kumar	kumar	PROPN
ijassa-348	256	2	:	:	PUNCT
ijassa-348	256	3	intrusion	intrusion	NOUN
ijassa-348	256	4	detection	detection	NOUN
ijassa-348	256	5	model	model	NOUN
ijassa-348	256	6	using	use	VERB
ijassa-348	256	7	pca	pca	PROPN
ijassa-348	256	8	and	and	CCONJ
ijassa-348	256	9	...	...	PUNCT
ijassa-348	256	10	where	where	SCONJ
ijassa-348	256	11	n=	n=	ADJ
ijassa-348	256	12	1	1	NUM
ijassa-348	256	13	·	·	PUNCT
ijassa-348	256	14	·	·	PUNCT
ijassa-348	256	15	·	·	PUNCT
ijassa-348	256	16	15	15	NUM
ijassa-348	256	17	for	for	ADP
ijassa-348	256	18	nsl	nsl	NOUN
ijassa-348	256	19	-	-	PUNCT
ijassa-348	256	20	kdd	kdd	NOUN
ijassa-348	256	21	dataset	dataset	NOUN
ijassa-348	256	22	and	and	CCONJ
ijassa-348	256	23	n=1	n=1	PROPN
ijassa-348	256	24	·	·	PUNCT
ijassa-348	256	25	·	·	PUNCT
ijassa-348	256	26	·	·	PUNCT
ijassa-348	256	27	27	27	NUM
ijassa-348	256	28	for	for	ADP
ijassa-348	256	29	unsw	unsw	NOUN
ijassa-348	256	30	-	-	PUNCT
ijassa-348	256	31	nb	nb	NOUN
ijassa-348	256	32	dataset	dataset	NOUN
ijassa-348	256	33	.	.	PUNCT
ijassa-348	257	1	each	each	DET
ijassa-348	257	2	coefficient	coefficient	NOUN
ijassa-348	257	3	wi	wi	PROPN
ijassa-348	257	4	is	be	AUX
ijassa-348	257	5	multiplied	multiply	VERB
ijassa-348	257	6	with	with	ADP
ijassa-348	257	7	output	output	NOUN
ijassa-348	257	8	from	from	ADP
ijassa-348	257	9	ith	ith	PROPN
ijassa-348	257	10	manager	manager	NOUN
ijassa-348	257	11	yi	yi	PROPN
ijassa-348	257	12	and	and	CCONJ
ijassa-348	257	13	the	the	DET
ijassa-348	257	14	final	final	ADJ
ijassa-348	257	15	decision	decision	NOUN
ijassa-348	257	16	is	be	AUX
ijassa-348	257	17	determined	determine	VERB
ijassa-348	257	18	by	by	ADP
ijassa-348	257	19	the	the	DET
ijassa-348	257	20	sign	sign	NOUN
ijassa-348	257	21	of	of	ADP
ijassa-348	257	22	the	the	DET
ijassa-348	257	23	sum	sum	NOUN
ijassa-348	257	24	of	of	ADP
ijassa-348	257	25	weight	weight	NOUN
ijassa-348	257	26	coefficients	coefficient	NOUN
ijassa-348	257	27	for	for	ADP
ijassa-348	257	28	all	all	DET
ijassa-348	257	29	managers	manager	NOUN
ijassa-348	257	30	.	.	PUNCT
ijassa-348	258	1	4	4	NUM
ijassa-348	258	2	proposed	propose	VERB
ijassa-348	258	3	model	model	NOUN
ijassa-348	258	4	the	the	DET
ijassa-348	258	5	proposed	propose	VERB
ijassa-348	258	6	model	model	NOUN
ijassa-348	258	7	is	be	AUX
ijassa-348	258	8	an	an	DET
ijassa-348	258	9	integrated	integrate	VERB
ijassa-348	258	10	intrusion	intrusion	NOUN
ijassa-348	258	11	detection	detection	NOUN
ijassa-348	258	12	system	system	NOUN
ijassa-348	258	13	combining	combine	VERB
ijassa-348	258	14	pca	pca	PROPN
ijassa-348	258	15	as	as	ADP
ijassa-348	258	16	the	the	DET
ijassa-348	258	17	dimensionality	dimensionality	NOUN
ijassa-348	258	18	reduction	reduction	NOUN
ijassa-348	258	19	technique	technique	NOUN
ijassa-348	258	20	and	and	CCONJ
ijassa-348	258	21	a	a	DET
ijassa-348	258	22	hybrid	hybrid	ADJ
ijassa-348	258	23	model	model	NOUN
ijassa-348	258	24	of	of	ADP
ijassa-348	258	25	base	base	NOUN
ijassa-348	258	26	and	and	CCONJ
ijassa-348	258	27	ensemble	ensemble	ADJ
ijassa-348	258	28	classifiers	classifier	NOUN
ijassa-348	258	29	.	.	PUNCT
ijassa-348	259	1	in	in	ADP
ijassa-348	259	2	stage	stage	NOUN
ijassa-348	259	3	1	1	NUM
ijassa-348	259	4	,	,	PUNCT
ijassa-348	259	5	the	the	DET
ijassa-348	259	6	raw	raw	ADJ
ijassa-348	259	7	data	data	NOUN
ijassa-348	259	8	is	be	AUX
ijassa-348	259	9	sent	send	VERB
ijassa-348	259	10	to	to	ADP
ijassa-348	259	11	a	a	DET
ijassa-348	259	12	preprocessing	preprocessing	NOUN
ijassa-348	259	13	unit	unit	NOUN
ijassa-348	259	14	for	for	ADP
ijassa-348	259	15	performing	perform	VERB
ijassa-348	259	16	normalization	normalization	NOUN
ijassa-348	259	17	by	by	ADP
ijassa-348	259	18	z	z	NOUN
ijassa-348	259	19	-	-	PUNCT
ijassa-348	259	20	score	score	NOUN
ijassa-348	259	21	technique	technique	NOUN
ijassa-348	259	22	and	and	CCONJ
ijassa-348	259	23	the	the	DET
ijassa-348	259	24	noisy	noisy	ADJ
ijassa-348	259	25	attributes	attribute	NOUN
ijassa-348	259	26	are	be	AUX
ijassa-348	259	27	removed	remove	VERB
ijassa-348	259	28	using	use	VERB
ijassa-348	259	29	dimensionality	dimensionality	NOUN
ijassa-348	259	30	reduction	reduction	NOUN
ijassa-348	259	31	.	.	PUNCT
ijassa-348	260	1	the	the	DET
ijassa-348	260	2	resultant	resultant	NOUN
ijassa-348	260	3	subset	subset	VERB
ijassa-348	260	4	is	be	AUX
ijassa-348	260	5	then	then	ADV
ijassa-348	260	6	fed	feed	VERB
ijassa-348	260	7	to	to	ADP
ijassa-348	260	8	the	the	DET
ijassa-348	260	9	stage	stage	NOUN
ijassa-348	260	10	2	2	NUM
ijassa-348	260	11	which	which	PRON
ijassa-348	260	12	is	be	AUX
ijassa-348	260	13	the	the	DET
ijassa-348	260	14	ensemble	ensemble	ADJ
ijassa-348	260	15	layer	layer	NOUN
ijassa-348	260	16	for	for	ADP
ijassa-348	260	17	classification	classification	NOUN
ijassa-348	260	18	.	.	PUNCT
ijassa-348	261	1	three	three	NUM
ijassa-348	261	2	classifiers	classifier	NOUN
ijassa-348	261	3	are	be	AUX
ijassa-348	261	4	deployed	deploy	VERB
ijassa-348	261	5	in	in	ADP
ijassa-348	261	6	stage	stage	NOUN
ijassa-348	261	7	2	2	NUM
ijassa-348	261	8	namely	namely	ADV
ijassa-348	261	9	svm	svm	ADJ
ijassa-348	261	10	.	.	PROPN
ijassa-348	262	1	linear	linear	ADJ
ijassa-348	262	2	and	and	CCONJ
ijassa-348	262	3	quadratic	quadratic	ADJ
ijassa-348	262	4	discriminant	discriminant	ADJ
ijassa-348	262	5	classifiers	classifier	NOUN
ijassa-348	262	6	.	.	PUNCT
ijassa-348	263	1	thus	thus	ADV
ijassa-348	263	2	the	the	DET
ijassa-348	263	3	ensembled	ensemble	VERB
ijassa-348	263	4	approach	approach	NOUN
ijassa-348	263	5	is	be	AUX
ijassa-348	263	6	a	a	DET
ijassa-348	263	7	merger	merger	NOUN
ijassa-348	263	8	of	of	ADP
ijassa-348	263	9	different	different	ADJ
ijassa-348	263	10	supervised	supervised	ADJ
ijassa-348	263	11	classifiers	classifier	NOUN
ijassa-348	263	12	.	.	PUNCT
ijassa-348	264	1	a	a	DET
ijassa-348	264	2	5	5	NUM
ijassa-348	264	3	-	-	ADJ
ijassa-348	264	4	fold	fold	ADJ
ijassa-348	264	5	cross	cross	NOUN
ijassa-348	264	6	validation	validation	NOUN
ijassa-348	264	7	is	be	AUX
ijassa-348	264	8	performed	perform	VERB
ijassa-348	264	9	to	to	PART
ijassa-348	264	10	split	split	VERB
ijassa-348	264	11	the	the	DET
ijassa-348	264	12	data	datum	NOUN
ijassa-348	264	13	into	into	ADP
ijassa-348	264	14	training	training	NOUN
ijassa-348	264	15	and	and	CCONJ
ijassa-348	264	16	testing	testing	NOUN
ijassa-348	264	17	sets	set	NOUN
ijassa-348	264	18	.	.	PUNCT
ijassa-348	265	1	the	the	DET
ijassa-348	265	2	class	class	NOUN
ijassa-348	265	3	label	label	NOUN
ijassa-348	265	4	is	be	AUX
ijassa-348	265	5	obtained	obtain	VERB
ijassa-348	265	6	by	by	ADP
ijassa-348	265	7	majority	majority	NOUN
ijassa-348	265	8	voting	voting	NOUN
ijassa-348	265	9	from	from	ADP
ijassa-348	265	10	the	the	DET
ijassa-348	265	11	three	three	NUM
ijassa-348	265	12	classifier	classifier	NOUN
ijassa-348	265	13	results	result	NOUN
ijassa-348	265	14	.	.	PUNCT
ijassa-348	266	1	fig	fig	NOUN
ijassa-348	266	2	1	1	NUM
ijassa-348	266	3	depicts	depict	VERB
ijassa-348	266	4	the	the	DET
ijassa-348	266	5	proposed	propose	VERB
ijassa-348	266	6	intrusion	intrusion	NOUN
ijassa-348	266	7	detection	detection	NOUN
ijassa-348	266	8	model	model	NOUN
ijassa-348	266	9	.	.	PUNCT
ijassa-348	267	1	in	in	ADP
ijassa-348	267	2	the	the	DET
ijassa-348	267	3	next	next	ADJ
ijassa-348	267	4	subsection	subsection	NOUN
ijassa-348	267	5	we	we	PRON
ijassa-348	267	6	discuss	discuss	VERB
ijassa-348	267	7	the	the	DET
ijassa-348	267	8	approach	approach	NOUN
ijassa-348	267	9	for	for	ADP
ijassa-348	267	10	obtaining	obtain	VERB
ijassa-348	267	11	optimal	optimal	ADJ
ijassa-348	267	12	subset	subset	NOUN
ijassa-348	267	13	using	use	VERB
ijassa-348	267	14	pca	pca	PROPN
ijassa-348	267	15	and	and	CCONJ
ijassa-348	267	16	ensemble	ensemble	ADJ
ijassa-348	267	17	approach	approach	NOUN
ijassa-348	267	18	for	for	ADP
ijassa-348	267	19	classification	classification	NOUN
ijassa-348	267	20	of	of	ADP
ijassa-348	267	21	network	network	NOUN
ijassa-348	267	22	traffic	traffic	NOUN
ijassa-348	267	23	label	label	NOUN
ijassa-348	267	24	.	.	PUNCT
ijassa-348	268	1	algorithm	algorithm	NOUN
ijassa-348	268	2	for	for	ADP
ijassa-348	268	3	obtaining	obtain	VERB
ijassa-348	268	4	the	the	DET
ijassa-348	268	5	optimal	optimal	ADJ
ijassa-348	268	6	feature	feature	NOUN
ijassa-348	268	7	subset	subset	NOUN
ijassa-348	268	8	using	use	VERB
ijassa-348	268	9	pca	pca	PROPN
ijassa-348	268	10	:	:	PUNCT
ijassa-348	268	11	input(training	input(training	NOUN
ijassa-348	268	12	set	set	NOUN
ijassa-348	268	13	,	,	PUNCT
ijassa-348	268	14	test	test	NOUN
ijassa-348	268	15	set	set	NOUN
ijassa-348	268	16	)	)	PUNCT
ijassa-348	268	17	output(optimal	output(optimal	ADJ
ijassa-348	268	18	training	training	NOUN
ijassa-348	268	19	set	set	NOUN
ijassa-348	268	20	,	,	PUNCT
ijassa-348	268	21	optimal	optimal	ADJ
ijassa-348	268	22	test	test	NOUN
ijassa-348	268	23	set	set	NOUN
ijassa-348	268	24	)	)	PUNCT
ijassa-348	268	25	step	step	NOUN
ijassa-348	268	26	1	1	NUM
ijassa-348	268	27	:	:	PUNCT
ijassa-348	268	28	determine	determine	VERB
ijassa-348	268	29	the	the	DET
ijassa-348	268	30	size	size	NOUN
ijassa-348	268	31	of	of	ADP
ijassa-348	268	32	training	training	NOUN
ijassa-348	268	33	and	and	CCONJ
ijassa-348	268	34	test	test	NOUN
ijassa-348	268	35	data	datum	NOUN
ijassa-348	268	36	step	step	NOUN
ijassa-348	268	37	2	2	NUM
ijassa-348	268	38	:	:	PUNCT
ijassa-348	268	39	scale	scale	VERB
ijassa-348	268	40	the	the	DET
ijassa-348	268	41	training	training	NOUN
ijassa-348	268	42	and	and	CCONJ
ijassa-348	268	43	test	test	NOUN
ijassa-348	268	44	data	datum	NOUN
ijassa-348	268	45	step	step	NOUN
ijassa-348	268	46	3	3	NUM
ijassa-348	268	47	:	:	PUNCT
ijassa-348	268	48	subtract	subtract	VERB
ijassa-348	268	49	the	the	DET
ijassa-348	268	50	mean	mean	NOUN
ijassa-348	268	51	for	for	ADP
ijassa-348	268	52	each	each	DET
ijassa-348	268	53	row	row	NOUN
ijassa-348	268	54	m	m	VERB
ijassa-348	268	55	=	=	PUNCT
ijassa-348	269	1	∑n	∑n	NOUN
ijassa-348	269	2	k=1	k=1	PUNCT
ijassa-348	269	3	xk	xk	PROPN
ijassa-348	270	1	n	n	PROPN
ijassa-348	270	2	(	(	PUNCT
ijassa-348	270	3	16	16	NUM
ijassa-348	270	4	)	)	PUNCT
ijassa-348	270	5	wherein	wherein	SCONJ
ijassa-348	270	6	x	x	PUNCT
ijassa-348	270	7	specifies	specify	VERB
ijassa-348	270	8	the	the	DET
ijassa-348	270	9	individual	individual	ADJ
ijassa-348	270	10	elements	element	NOUN
ijassa-348	270	11	and	and	CCONJ
ijassa-348	270	12	‘	'	PUNCT
ijassa-348	270	13	n	n	CCONJ
ijassa-348	270	14	’	'	PUNCT
ijassa-348	270	15	denotes	denote	VERB
ijassa-348	270	16	the	the	DET
ijassa-348	270	17	no	no	NOUN
ijassa-348	270	18	.	.	PUNCT
ijassa-348	270	19	of	of	ADP
ijassa-348	270	20	samples	sample	NOUN
ijassa-348	270	21	.	.	PUNCT
ijassa-348	271	1	step	step	NOUN
ijassa-348	271	2	4	4	NUM
ijassa-348	271	3	:	:	PUNCT
ijassa-348	271	4	determine	determine	VERB
ijassa-348	271	5	the	the	DET
ijassa-348	271	6	covariance	covariance	NOUN
ijassa-348	271	7	matrix	matrix	NOUN
ijassa-348	271	8	c	c	NOUN
ijassa-348	272	1	=	=	SYM
ijassa-348	272	2	xixit	xixit	PROPN
ijassa-348	272	3	n	n	PROPN
ijassa-348	272	4	(	(	PUNCT
ijassa-348	272	5	17	17	NUM
ijassa-348	272	6	)	)	PUNCT
ijassa-348	272	7	where	where	SCONJ
ijassa-348	272	8	x	x	PRON
ijassa-348	272	9	represents	represent	VERB
ijassa-348	272	10	the	the	DET
ijassa-348	272	11	matrix	matrix	NOUN
ijassa-348	272	12	after	after	ADP
ijassa-348	272	13	subtracting	subtract	VERB
ijassa-348	272	14	the	the	DET
ijassa-348	272	15	mean	mean	NOUN
ijassa-348	272	16	and	and	CCONJ
ijassa-348	272	17	xt	xt	PROPN
ijassa-348	272	18	is	be	AUX
ijassa-348	272	19	the	the	DET
ijassa-348	272	20	transpose	transpose	ADJ
ijassa-348	272	21	matrix	matrix	NOUN
ijassa-348	272	22	and	and	CCONJ
ijassa-348	272	23	n	n	NOUN
ijassa-348	272	24	is	be	AUX
ijassa-348	272	25	the	the	DET
ijassa-348	272	26	total	total	ADJ
ijassa-348	272	27	number	number	NOUN
ijassa-348	272	28	of	of	ADP
ijassa-348	272	29	elements	element	NOUN
ijassa-348	272	30	.	.	PUNCT
ijassa-348	273	1	step	step	NOUN
ijassa-348	273	2	5	5	NUM
ijassa-348	273	3	:	:	PUNCT
ijassa-348	273	4	determine	determine	VERB
ijassa-348	273	5	the	the	DET
ijassa-348	273	6	eigenvectors	eigenvector	NOUN
ijassa-348	273	7	and	and	CCONJ
ijassa-348	273	8	eigenvalues	eigenvalue	NOUN
ijassa-348	273	9	of	of	ADP
ijassa-348	273	10	the	the	DET
ijassa-348	273	11	covariance	covariance	NOUN
ijassa-348	273	12	matrix	matrix	NOUN
ijassa-348	273	13	.	.	PUNCT
ijassa-348	274	1	σv	σv	NOUN
ijassa-348	274	2	=	=	SYM
ijassa-348	275	1	λv	λv	X
ijassa-348	275	2	(	(	PUNCT
ijassa-348	275	3	18	18	NUM
ijassa-348	275	4	)	)	PUNCT
ijassa-348	275	5	step	step	NOUN
ijassa-348	275	6	6	6	NUM
ijassa-348	275	7	:	:	PUNCT
ijassa-348	275	8	obtain	obtain	VERB
ijassa-348	275	9	a	a	DET
ijassa-348	275	10	feature	feature	NOUN
ijassa-348	275	11	vector	vector	NOUN
ijassa-348	275	12	=	=	PUNCT
ijassa-348	275	13	(	(	PUNCT
ijassa-348	275	14	eig1	eig1	PROPN
ijassa-348	275	15	,	,	PUNCT
ijassa-348	275	16	eig2	eig2	PROPN
ijassa-348	275	17	...	...	PUNCT
ijassa-348	275	18	eigp	eigp	NOUN
ijassa-348	275	19	)	)	PUNCT
ijassa-348	275	20	where	where	SCONJ
ijassa-348	275	21	eig1	eig1	PROPN
ijassa-348	275	22	is	be	AUX
ijassa-348	275	23	principal	principal	ADJ
ijassa-348	275	24	component	component	NOUN
ijassa-348	275	25	and	and	CCONJ
ijassa-348	276	1	p	p	NOUN
ijassa-348	276	2	≤	≤	PROPN
ijassa-348	276	3	n.	n.	NOUN
ijassa-348	276	4	select	select	ADJ
ijassa-348	276	5	’	'	PUNCT
ijassa-348	276	6	m	m	NOUN
ijassa-348	276	7	’	'	PUNCT
ijassa-348	276	8	such	such	ADJ
ijassa-348	276	9	eigen	eigen	PROPN
ijassa-348	276	10	vectors	vector	NOUN
ijassa-348	276	11	that	that	PRON
ijassa-348	276	12	match	match	VERB
ijassa-348	276	13	to	to	ADP
ijassa-348	276	14	the	the	DET
ijassa-348	276	15	largest	large	ADJ
ijassa-348	276	16	advances	advance	NOUN
ijassa-348	276	17	in	in	ADP
ijassa-348	276	18	systems	system	NOUN
ijassa-348	276	19	science	science	NOUN
ijassa-348	276	20	and	and	CCONJ
ijassa-348	276	21	application	application	NOUN
ijassa-348	276	22	(	(	PUNCT
ijassa-348	276	23	2016	2016	NUM
ijassa-348	276	24	)	)	PUNCT
ijassa-348	277	1	vol.16	vol.16	PROPN
ijassa-348	277	2	no.2	no.2	PROPN
ijassa-348	277	3	27	27	NUM
ijassa-348	277	4	‘	'	PUNCT
ijassa-348	277	5	m	m	PRON
ijassa-348	277	6	’	'	PUNCT
ijassa-348	277	7	eigenvalues	eigenvalue	VERB
ijassa-348	277	8	in	in	ADP
ijassa-348	277	9	the	the	DET
ijassa-348	277	10	set	set	NOUN
ijassa-348	277	11	.	.	PUNCT
ijassa-348	278	1	algorithm	algorithm	NOUN
ijassa-348	278	2	for	for	ADP
ijassa-348	278	3	obtaining	obtain	VERB
ijassa-348	278	4	the	the	DET
ijassa-348	278	5	class	class	NOUN
ijassa-348	278	6	label	label	NOUN
ijassa-348	278	7	using	use	VERB
ijassa-348	278	8	hybrid	hybrid	ADJ
ijassa-348	278	9	model	model	NOUN
ijassa-348	278	10	given	give	VERB
ijassa-348	278	11	:	:	PUNCT
ijassa-348	278	12	classifier	classifier	NOUN
ijassa-348	278	13	m1,(svm)m2,(ldc),m3(qdc	m1,(svm)m2,(ldc),m3(qdc	NOUN
ijassa-348	278	14	)	)	PUNCT
ijassa-348	278	15	,	,	PUNCT
ijassa-348	278	16	ensemble(wma	ensemble(wma	NOUN
ijassa-348	278	17	)	)	PUNCT
ijassa-348	278	18	input	input	NOUN
ijassa-348	278	19	:	:	PUNCT
ijassa-348	278	20	optimal	optimal	ADJ
ijassa-348	278	21	attribute	attribute	NOUN
ijassa-348	278	22	dataset	dataset	VERB
ijassa-348	278	23	d	d	X
ijassa-348	278	24	output	output	NOUN
ijassa-348	278	25	:	:	PUNCT
ijassa-348	278	26	class	class	NOUN
ijassa-348	278	27	label	label	NOUN
ijassa-348	278	28	step1	step1	PROPN
ijassa-348	278	29	:	:	PUNCT
ijassa-348	278	30	initialize	initialize	VERB
ijassa-348	278	31	all	all	DET
ijassa-348	278	32	the	the	DET
ijassa-348	278	33	weights	weight	NOUN
ijassa-348	278	34	in	in	ADP
ijassa-348	278	35	d.	d.	PROPN
ijassa-348	278	36	wi	wi	PROPN
ijassa-348	278	37	=	=	PROPN
ijassa-348	278	38	1	1	NUM
ijassa-348	278	39	/	/	SYM
ijassa-348	278	40	n	n	CCONJ
ijassa-348	278	41	,	,	PUNCT
ijassa-348	278	42	where	where	SCONJ
ijassa-348	278	43	n	n	X
ijassa-348	278	44	is	be	AUX
ijassa-348	278	45	the	the	DET
ijassa-348	278	46	total	total	ADJ
ijassa-348	278	47	number	number	NOUN
ijassa-348	278	48	of	of	ADP
ijassa-348	278	49	elements	element	NOUN
ijassa-348	278	50	.	.	PUNCT
ijassa-348	279	1	step2	step2	PROPN
ijassa-348	279	2	:	:	PUNCT
ijassa-348	279	3	for	for	SCONJ
ijassa-348	279	4	every	every	DET
ijassa-348	279	5	sample	sample	NOUN
ijassa-348	279	6	data	datum	NOUN
ijassa-348	279	7	di	di	NOUN
ijassa-348	279	8	fit	fit	VERB
ijassa-348	279	9	the	the	DET
ijassa-348	279	10	svm	svm	ADJ
ijassa-348	279	11	classifier	classifier	NOUN
ijassa-348	279	12	to	to	ADP
ijassa-348	279	13	(	(	PUNCT
ijassa-348	279	14	xt	xt	PROPN
ijassa-348	279	15	,	,	PUNCT
ijassa-348	279	16	yt	yt	PROPN
ijassa-348	279	17	)	)	PUNCT
ijassa-348	279	18	using	use	VERB
ijassa-348	279	19	weights	weight	NOUN
ijassa-348	279	20	wi	wi	PROPN
ijassa-348	279	21	for	for	ADP
ijassa-348	279	22	each	each	DET
ijassa-348	279	23	class	class	NOUN
ijassa-348	279	24	label	label	NOUN
ijassa-348	279	25	k	k	NOUN
ijassa-348	279	26	=	=	NOUN
ijassa-348	279	27	1	1	NUM
ijassa-348	279	28	...	...	PUNCT
ijassa-348	279	29	k	k	X
ijassa-348	279	30	obtain	obtain	VERB
ijassa-348	279	31	the	the	DET
ijassa-348	279	32	hypothesis	hypothesis	NOUN
ijassa-348	279	33	xs	xs	PROPN
ijassa-348	279	34	←	←	PROPN
ijassa-348	279	35	argmin	argmin	PROPN
ijassa-348	279	36	(	(	PUNCT
ijassa-348	279	37	λ	λ	X
ijassa-348	279	38	|	|	NOUN
ijassa-348	279	39	f	f	PROPN
ijassa-348	279	40	(	(	PUNCT
ijassa-348	279	41	xj	xj	PROPN
ijassa-348	279	42	)	)	PUNCT
ijassa-348	280	1	|	|	ADV
ijassa-348	280	2	+	+	PROPN
ijassa-348	280	3	(	(	PUNCT
ijassa-348	280	4	1−	1−	NUM
ijassa-348	280	5	λ	λ	NOUN
ijassa-348	280	6	)	)	PUNCT
ijassa-348	280	7	)	)	PUNCT
ijassa-348	280	8	max	max	PROPN
ijassa-348	280	9	k(xi	k(xi	PROPN
ijassa-348	280	10	,	,	PUNCT
ijassa-348	280	11	xj)√	xj)√	PUNCT
ijassa-348	281	1	k(xi	k(xi	PROPN
ijassa-348	281	2	,	,	PUNCT
ijassa-348	281	3	xj)k(xi	xj)k(xi	PROPN
ijassa-348	281	4	,	,	PUNCT
ijassa-348	281	5	xj	xj	PROPN
ijassa-348	281	6	)	)	PUNCT
ijassa-348	282	1	d	d	X
ijassa-348	282	2	←	←	PROPN
ijassa-348	282	3	d	d	X
ijassa-348	282	4	∪	∪	X
ijassa-348	282	5	{	{	PUNCT
ijassa-348	282	6	xs	xs	NOUN
ijassa-348	282	7	}	}	PUNCT
ijassa-348	282	8	label(d	label(d	PROPN
ijassa-348	282	9	)	)	PUNCT
ijassa-348	282	10	l←	l←	PUNCT
ijassa-348	282	11	l	l	NOUN
ijassa-348	282	12	∪	∪	X
ijassa-348	282	13	{	{	PUNCT
ijassa-348	282	14	s	s	NOUN
ijassa-348	282	15	}	}	PUNCT
ijassa-348	282	16	step	step	NOUN
ijassa-348	282	17	3	3	NUM
ijassa-348	282	18	:	:	PUNCT
ijassa-348	282	19	for	for	SCONJ
ijassa-348	282	20	every	every	DET
ijassa-348	282	21	sample	sample	NOUN
ijassa-348	282	22	data	datum	NOUN
ijassa-348	282	23	di	di	PROPN
ijassa-348	282	24	i	i	PROPN
ijassa-348	282	25	)	)	PUNCT
ijassa-348	282	26	compute	compute	VERB
ijassa-348	282	27	the	the	DET
ijassa-348	282	28	sample	sample	NOUN
ijassa-348	282	29	estimates	estimate	VERB
ijassa-348	282	30	π̂m	π̂m	PRON
ijassa-348	282	31	,	,	PUNCT
ijassa-348	282	32	ûm	ûm	PROPN
ijassa-348	282	33	,	,	PUNCT
ijassa-348	282	34	σ̂	σ̂	PROPN
ijassa-348	282	35	ii	ii	NOUN
ijassa-348	282	36	)	)	PUNCT
ijassa-348	282	37	make	make	VERB
ijassa-348	282	38	two	two	NUM
ijassa-348	282	39	transformations	transformation	NOUN
ijassa-348	282	40	:	:	PUNCT
ijassa-348	283	1	sphere	sphere	VERB
ijassa-348	283	2	the	the	DET
ijassa-348	283	3	data	data	NOUN
ijassa-348	283	4	points	point	NOUN
ijassa-348	283	5	based	base	VERB
ijassa-348	283	6	on	on	ADP
ijassa-348	283	7	factoring	factor	VERB
ijassa-348	283	8	σ̂	σ̂	PROPN
ijassa-348	283	9	and	and	CCONJ
ijassa-348	283	10	project	project	NOUN
ijassa-348	283	11	to	to	ADP
ijassa-348	283	12	the	the	DET
ijassa-348	283	13	subspace	subspace	NOUN
ijassa-348	283	14	by	by	ADP
ijassa-348	283	15	the	the	DET
ijassa-348	283	16	centroids	centroid	NOUN
ijassa-348	283	17	.	.	PUNCT
ijassa-348	284	1	thus	thus	ADV
ijassa-348	284	2	a	a	DET
ijassa-348	284	3	transformation	transformation	NOUN
ijassa-348	284	4	of	of	ADP
ijassa-348	284	5	aϵrp(k−1	aϵrp(k−1	PROPN
ijassa-348	284	6	)	)	PUNCT
ijassa-348	284	7	iii	iii	NOUN
ijassa-348	284	8	)	)	PUNCT
ijassa-348	284	9	given	give	VERB
ijassa-348	284	10	any	any	DET
ijassa-348	284	11	data	datum	NOUN
ijassa-348	284	12	,	,	PUNCT
ijassa-348	284	13	xϵrp	xϵrp	PROPN
ijassa-348	284	14	transform	transform	VERB
ijassa-348	284	15	to	to	ADP
ijassa-348	284	16	x̄	x̄	NOUN
ijassa-348	284	17	=	=	PUNCT
ijassa-348	284	18	axϵrk−1	axϵrk−1	PROPN
ijassa-348	284	19	x̄	x̄	PUNCT
ijassa-348	284	20	=	=	SYM
ijassa-348	284	21	axϵrk−1	axϵrk−1	PROPN
ijassa-348	284	22	and	and	CCONJ
ijassa-348	284	23	classify	classify	VERB
ijassa-348	284	24	according	accord	VERB
ijassa-348	284	25	to	to	ADP
ijassa-348	284	26	class	class	NOUN
ijassa-348	284	27	m	m	NOUN
ijassa-348	284	28	=	=	NOUN
ijassa-348	284	29	1	1	NUM
ijassa-348	284	30	...	...	PUNCT
ijassa-348	284	31	k	k	NOUN
ijassa-348	284	32	for	for	ADP
ijassa-348	284	33	which	which	PRON
ijassa-348	284	34	1	1	NUM
ijassa-348	284	35	2∥	2∥	NUM
ijassa-348	284	36	x̃−	x̃−	PROPN
ijassa-348	284	37	ũm	ũm	PROPN
ijassa-348	285	1	∥2	∥2	PRON
ijassa-348	285	2	−	−	NOUN
ijassa-348	285	3	log	log	NOUN
ijassa-348	285	4	π̂m	π̂m	VERB
ijassa-348	285	5	is	be	AUX
ijassa-348	285	6	lowest	low	ADJ
ijassa-348	285	7	where	where	SCONJ
ijassa-348	285	8	ũj	ũj	ADJ
ijassa-348	285	9	=	=	SYM
ijassa-348	285	10	aûj	aûj	PROPN
ijassa-348	285	11	step	step	NOUN
ijassa-348	285	12	4	4	NUM
ijassa-348	285	13	:	:	PUNCT
ijassa-348	285	14	for	for	ADP
ijassa-348	285	15	every	every	DET
ijassa-348	285	16	sample	sample	NOUN
ijassa-348	285	17	data	datum	NOUN
ijassa-348	285	18	di	di	PROPN
ijassa-348	285	19	,	,	PUNCT
ijassa-348	285	20	perform	perform	VERB
ijassa-348	285	21	qda	qda	NOUN
ijassa-348	285	22	by	by	ADP
ijassa-348	285	23	repeating	repeat	VERB
ijassa-348	285	24	the	the	DET
ijassa-348	285	25	process	process	NOUN
ijassa-348	285	26	in	in	ADP
ijassa-348	285	27	step	step	NOUN
ijassa-348	285	28	3	3	NUM
ijassa-348	285	29	but	but	CCONJ
ijassa-348	285	30	with	with	ADP
ijassa-348	285	31	different	different	ADJ
ijassa-348	285	32	and	and	CCONJ
ijassa-348	285	33	for	for	ADP
ijassa-348	285	34	each	each	DET
ijassa-348	285	35	class	class	NOUN
ijassa-348	285	36	and	and	CCONJ
ijassa-348	285	37	compute	compute	VERB
ijassa-348	285	38	the	the	DET
ijassa-348	285	39	class	class	NOUN
ijassa-348	285	40	label	label	NOUN
ijassa-348	285	41	.	.	PUNCT
ijassa-348	286	1	step	step	NOUN
ijassa-348	286	2	5	5	NUM
ijassa-348	286	3	:	:	PUNCT
ijassa-348	286	4	the	the	DET
ijassa-348	286	5	class	class	NOUN
ijassa-348	286	6	label	label	NOUN
ijassa-348	286	7	is	be	AUX
ijassa-348	286	8	predicted	predict	VERB
ijassa-348	286	9	by	by	ADP
ijassa-348	286	10	weighted	weight	VERB
ijassa-348	286	11	majority	majority	NOUN
ijassa-348	286	12	voting	voting	NOUN
ijassa-348	286	13	from	from	ADP
ijassa-348	286	14	results	result	NOUN
ijassa-348	286	15	of	of	ADP
ijassa-348	286	16	steps	step	NOUN
ijassa-348	286	17	(	(	PUNCT
ijassa-348	286	18	2	2	NUM
ijassa-348	286	19	)	)	PUNCT
ijassa-348	286	20	,	,	PUNCT
ijassa-348	286	21	(	(	PUNCT
ijassa-348	286	22	3)and	3)and	NUM
ijassa-348	286	23	(	(	PUNCT
ijassa-348	286	24	4	4	NUM
ijassa-348	286	25	)	)	PUNCT
ijassa-348	286	26	given	give	VERB
ijassa-348	286	27	in	in	ADP
ijassa-348	286	28	equation	equation	NOUN
ijassa-348	286	29	(	(	PUNCT
ijassa-348	286	30	15	15	NUM
ijassa-348	286	31	)	)	PUNCT
ijassa-348	286	32	step	step	NOUN
ijassa-348	286	33	6	6	NUM
ijassa-348	286	34	:	:	PUNCT
ijassa-348	286	35	determine	determine	VERB
ijassa-348	286	36	the	the	DET
ijassa-348	286	37	performance	performance	NOUN
ijassa-348	286	38	of	of	ADP
ijassa-348	286	39	the	the	DET
ijassa-348	286	40	model	model	NOUN
ijassa-348	286	41	and	and	CCONJ
ijassa-348	286	42	analyze	analyze	VERB
ijassa-348	286	43	the	the	DET
ijassa-348	286	44	accuracy	accuracy	NOUN
ijassa-348	286	45	and	and	CCONJ
ijassa-348	286	46	time	time	NOUN
ijassa-348	286	47	complexity	complexity	NOUN
ijassa-348	286	48	of	of	ADP
ijassa-348	286	49	the	the	DET
ijassa-348	286	50	different	different	ADJ
ijassa-348	286	51	algorithms	algorithm	NOUN
ijassa-348	286	52	.	.	PUNCT
ijassa-348	287	1	4.1	4.1	NUM
ijassa-348	287	2	class	class	NOUN
ijassa-348	287	3	label	label	NOUN
ijassa-348	287	4	prediction	prediction	NOUN
ijassa-348	287	5	using	use	VERB
ijassa-348	287	6	weighted	weight	VERB
ijassa-348	287	7	majority	majority	NOUN
ijassa-348	287	8	approach	approach	NOUN
ijassa-348	287	9	(	(	PUNCT
ijassa-348	287	10	wma	wma	NOUN
ijassa-348	287	11	)	)	PUNCT
ijassa-348	287	12	in	in	ADP
ijassa-348	287	13	this	this	DET
ijassa-348	287	14	paper	paper	NOUN
ijassa-348	287	15	,	,	PUNCT
ijassa-348	287	16	we	we	PRON
ijassa-348	287	17	first	first	ADV
ijassa-348	287	18	deploy	deploy	VERB
ijassa-348	287	19	svm	svm	PROPN
ijassa-348	287	20	,	,	PUNCT
ijassa-348	287	21	ldc	ldc	PROPN
ijassa-348	287	22	and	and	CCONJ
ijassa-348	287	23	qdc	qdc	VERB
ijassa-348	287	24	classifiers	classifier	NOUN
ijassa-348	287	25	individually	individually	ADV
ijassa-348	287	26	to	to	PART
ijassa-348	287	27	result	result	VERB
ijassa-348	287	28	in	in	ADP
ijassa-348	287	29	a	a	DET
ijassa-348	287	30	good	good	ADJ
ijassa-348	287	31	generalization	generalization	NOUN
ijassa-348	287	32	performance	performance	NOUN
ijassa-348	287	33	.	.	PUNCT
ijassa-348	288	1	after	after	ADP
ijassa-348	288	2	passing	pass	VERB
ijassa-348	288	3	through	through	ADP
ijassa-348	288	4	each	each	PRON
ijassa-348	288	5	of	of	ADP
ijassa-348	288	6	the	the	DET
ijassa-348	288	7	individual	individual	ADJ
ijassa-348	288	8	classifiers	classifier	NOUN
ijassa-348	288	9	,	,	PUNCT
ijassa-348	288	10	a	a	DET
ijassa-348	288	11	majority	majority	NOUN
ijassa-348	288	12	voting	voting	NOUN
ijassa-348	288	13	technique	technique	NOUN
ijassa-348	288	14	is	be	AUX
ijassa-348	288	15	adopted	adopt	VERB
ijassa-348	288	16	to	to	PART
ijassa-348	288	17	predict	predict	VERB
ijassa-348	288	18	the	the	DET
ijassa-348	288	19	resultant	resultant	NOUN
ijassa-348	288	20	class	class	NOUN
ijassa-348	288	21	label	label	NOUN
ijassa-348	288	22	from	from	ADP
ijassa-348	288	23	the	the	DET
ijassa-348	288	24	different	different	ADJ
ijassa-348	288	25	classifier	classifier	NOUN
ijassa-348	288	26	labels	label	NOUN
ijassa-348	288	27	.	.	PUNCT
ijassa-348	289	1	the	the	DET
ijassa-348	289	2	major	major	ADJ
ijassa-348	289	3	advantage	advantage	NOUN
ijassa-348	289	4	of	of	ADP
ijassa-348	289	5	using	use	VERB
ijassa-348	289	6	ensemble	ensemble	ADJ
ijassa-348	289	7	approach	approach	NOUN
ijassa-348	289	8	is	be	AUX
ijassa-348	289	9	that	that	SCONJ
ijassa-348	289	10	the	the	DET
ijassa-348	289	11	performance	performance	NOUN
ijassa-348	289	12	is	be	AUX
ijassa-348	289	13	improved	improve	VERB
ijassa-348	289	14	because	because	SCONJ
ijassa-348	289	15	the	the	DET
ijassa-348	289	16	approach	approach	NOUN
ijassa-348	289	17	selects	select	VERB
ijassa-348	289	18	only	only	ADV
ijassa-348	289	19	the	the	DET
ijassa-348	289	20	class	class	NOUN
ijassa-348	289	21	label	label	NOUN
ijassa-348	289	22	which	which	PRON
ijassa-348	289	23	are	be	AUX
ijassa-348	289	24	correctly	correctly	ADV
ijassa-348	289	25	identified	identify	VERB
ijassa-348	289	26	by	by	ADP
ijassa-348	289	27	all	all	DET
ijassa-348	289	28	the	the	DET
ijassa-348	289	29	classification	classification	NOUN
ijassa-348	289	30	techniques	technique	NOUN
ijassa-348	289	31	.	.	PUNCT
ijassa-348	290	1	the	the	DET
ijassa-348	290	2	efficiency	efficiency	NOUN
ijassa-348	290	3	of	of	ADP
ijassa-348	290	4	existing	exist	VERB
ijassa-348	290	5	approaches	approach	NOUN
ijassa-348	290	6	is	be	AUX
ijassa-348	290	7	compared	compare	VERB
ijassa-348	290	8	by	by	ADP
ijassa-348	290	9	deploying	deploy	VERB
ijassa-348	290	10	a	a	DET
ijassa-348	290	11	validation	validation	NOUN
ijassa-348	290	12	set	set	VERB
ijassa-348	290	13	to	to	PART
ijassa-348	290	14	obtain	obtain	VERB
ijassa-348	290	15	weights	weight	NOUN
ijassa-348	290	16	for	for	ADP
ijassa-348	290	17	each	each	DET
ijassa-348	290	18	classifier	classifier	NOUN
ijassa-348	290	19	.	.	PUNCT
ijassa-348	291	1	all	all	DET
ijassa-348	291	2	managers	manager	NOUN
ijassa-348	291	3	assign	assign	VERB
ijassa-348	291	4	initial	initial	ADJ
ijassa-348	291	5	weight	weight	NOUN
ijassa-348	291	6	with	with	ADP
ijassa-348	291	7	value	value	NOUN
ijassa-348	291	8	1	1	NUM
ijassa-348	291	9	and	and	CCONJ
ijassa-348	291	10	all	all	DET
ijassa-348	291	11	the	the	DET
ijassa-348	291	12	managers	manager	NOUN
ijassa-348	291	13	weights	weight	VERB
ijassa-348	291	14	28	28	NUM
ijassa-348	291	15	sumaiya	sumaiya	PROPN
ijassa-348	291	16	thaseen	thaseen	PROPN
ijassa-348	291	17	and	and	CCONJ
ijassa-348	291	18	ch.aswani	ch.aswani	X
ijassa-348	292	1	kumar	kumar	PROPN
ijassa-348	292	2	:	:	PUNCT
ijassa-348	292	3	intrusion	intrusion	NOUN
ijassa-348	292	4	detection	detection	NOUN
ijassa-348	292	5	model	model	NOUN
ijassa-348	292	6	using	use	VERB
ijassa-348	292	7	pca	pca	PROPN
ijassa-348	292	8	and	and	CCONJ
ijassa-348	292	9	...	...	PUNCT
ijassa-348	292	10	are	be	AUX
ijassa-348	292	11	combined	combine	VERB
ijassa-348	292	12	using	use	VERB
ijassa-348	292	13	wmv	wmv	NOUN
ijassa-348	292	14	.	.	PUNCT
ijassa-348	293	1	the	the	DET
ijassa-348	293	2	predicted	predict	VERB
ijassa-348	293	3	class	class	NOUN
ijassa-348	293	4	label	label	NOUN
ijassa-348	293	5	is	be	AUX
ijassa-348	293	6	then	then	ADV
ijassa-348	293	7	compared	compare	VERB
ijassa-348	293	8	with	with	ADP
ijassa-348	293	9	the	the	DET
ijassa-348	293	10	target	target	NOUN
ijassa-348	293	11	label	label	NOUN
ijassa-348	293	12	present	present	ADJ
ijassa-348	293	13	in	in	ADP
ijassa-348	293	14	the	the	DET
ijassa-348	293	15	validation	validation	NOUN
ijassa-348	293	16	set	set	NOUN
ijassa-348	293	17	.	.	PUNCT
ijassa-348	294	1	if	if	SCONJ
ijassa-348	294	2	the	the	DET
ijassa-348	294	3	manager	manager	NOUN
ijassa-348	294	4	has	have	AUX
ijassa-348	294	5	made	make	VERB
ijassa-348	294	6	a	a	DET
ijassa-348	294	7	mistake	mistake	NOUN
ijassa-348	294	8	in	in	ADP
ijassa-348	294	9	identifying	identify	VERB
ijassa-348	294	10	the	the	DET
ijassa-348	294	11	class	class	NOUN
ijassa-348	294	12	label	label	NOUN
ijassa-348	294	13	then	then	ADV
ijassa-348	294	14	the	the	DET
ijassa-348	294	15	weights	weight	NOUN
ijassa-348	294	16	will	will	AUX
ijassa-348	294	17	be	be	AUX
ijassa-348	294	18	subtracted	subtract	VERB
ijassa-348	294	19	by	by	ADP
ijassa-348	294	20	a	a	DET
ijassa-348	294	21	learning	learn	VERB
ijassa-348	294	22	factor	factor	NOUN
ijassa-348	294	23	β	β	NOUN
ijassa-348	294	24	.	.	PUNCT
ijassa-348	295	1	learning	learn	VERB
ijassa-348	295	2	factor	factor	NOUN
ijassa-348	295	3	is	be	AUX
ijassa-348	295	4	user	user	NOUN
ijassa-348	295	5	defined	define	VERB
ijassa-348	295	6	and	and	CCONJ
ijassa-348	295	7	the	the	DET
ijassa-348	295	8	values	value	NOUN
ijassa-348	295	9	range	range	VERB
ijassa-348	295	10	between	between	ADP
ijassa-348	295	11	0	0	NUM
ijassa-348	295	12	and	and	CCONJ
ijassa-348	295	13	1	1	NUM
ijassa-348	295	14	.	.	PUNCT
ijassa-348	296	1	this	this	DET
ijassa-348	296	2	process	process	NOUN
ijassa-348	296	3	is	be	AUX
ijassa-348	296	4	repeated	repeat	VERB
ijassa-348	296	5	for	for	ADP
ijassa-348	296	6	every	every	DET
ijassa-348	296	7	sample	sample	NOUN
ijassa-348	296	8	in	in	ADP
ijassa-348	296	9	the	the	DET
ijassa-348	296	10	dataset	dataset	NOUN
ijassa-348	296	11	.	.	PUNCT
ijassa-348	297	1	after	after	ADP
ijassa-348	297	2	each	each	DET
ijassa-348	297	3	test	test	NOUN
ijassa-348	297	4	in	in	ADP
ijassa-348	297	5	which	which	PRON
ijassa-348	297	6	a	a	DET
ijassa-348	297	7	mistake	mistake	NOUN
ijassa-348	297	8	arises	arise	VERB
ijassa-348	297	9	the	the	DET
ijassa-348	297	10	sum	sum	NOUN
ijassa-348	297	11	of	of	ADP
ijassa-348	297	12	the	the	DET
ijassa-348	297	13	weights	weight	NOUN
ijassa-348	297	14	is	be	AUX
ijassa-348	297	15	at	at	ADP
ijassa-348	297	16	most	most	ADJ
ijassa-348	297	17	‘	'	PUNCT
ijassa-348	297	18	u	u	NOUN
ijassa-348	297	19	’	'	PUNCT
ijassa-348	297	20	times	time	NOUN
ijassa-348	297	21	the	the	DET
ijassa-348	297	22	sum	sum	NOUN
ijassa-348	297	23	of	of	ADP
ijassa-348	297	24	the	the	DET
ijassa-348	297	25	weights	weight	NOUN
ijassa-348	297	26	before	before	ADP
ijassa-348	297	27	the	the	DET
ijassa-348	297	28	test	test	NOUN
ijassa-348	297	29	wstart	wstart	NOUN
ijassa-348	297	30	for	for	ADP
ijassa-348	297	31	specific	specific	ADJ
ijassa-348	297	32	u	u	NOUN
ijassa-348	297	33	<	<	X
ijassa-348	297	34	1	1	NUM
ijassa-348	297	35	.	.	PUNCT
ijassa-348	298	1	if	if	SCONJ
ijassa-348	298	2	the	the	DET
ijassa-348	298	3	initial	initial	ADJ
ijassa-348	298	4	eight	eight	NUM
ijassa-348	298	5	is	be	AUX
ijassa-348	298	6	wbegin	wbegin	NOUN
ijassa-348	298	7	and	and	CCONJ
ijassa-348	298	8	the	the	DET
ijassa-348	298	9	final	final	ADJ
ijassa-348	298	10	weight	weight	NOUN
ijassa-348	298	11	is	be	AUX
ijassa-348	298	12	wfin	wfin	NOUN
ijassa-348	298	13	,	,	PUNCT
ijassa-348	298	14	then	then	ADV
ijassa-348	298	15	wbeginu	wbeginu	VERB
ijassa-348	298	16	f	f	PROPN
ijassa-348	298	17	≥	≥	NOUN
ijassa-348	298	18	wfin	wfin	NOUN
ijassa-348	298	19	must	must	AUX
ijassa-348	298	20	be	be	AUX
ijassa-348	298	21	true	true	ADJ
ijassa-348	298	22	where	where	SCONJ
ijassa-348	298	23	‘	'	PUNCT
ijassa-348	298	24	f	f	X
ijassa-348	298	25	’	'	PUNCT
ijassa-348	298	26	is	be	AUX
ijassa-348	298	27	the	the	DET
ijassa-348	298	28	number	number	NOUN
ijassa-348	298	29	of	of	ADP
ijassa-348	298	30	faults	fault	NOUN
ijassa-348	298	31	.	.	PUNCT
ijassa-348	299	1	f	f	PROPN
ijassa-348	299	2	≤	≤	PROPN
ijassa-348	299	3	log	log	NOUN
ijassa-348	299	4	(	(	PUNCT
ijassa-348	299	5	wbegin	wbegin	NOUN
ijassa-348	299	6	−wfin	−wfin	NOUN
ijassa-348	299	7	)	)	PUNCT
ijassa-348	299	8	log	log	NOUN
ijassa-348	299	9	(	(	PUNCT
ijassa-348	299	10	1	1	NUM
ijassa-348	299	11	/	/	SYM
ijassa-348	299	12	u	u	NOUN
ijassa-348	299	13	)	)	PUNCT
ijassa-348	299	14	(	(	PUNCT
ijassa-348	299	15	19	19	NUM
ijassa-348	299	16	)	)	PUNCT
ijassa-348	299	17	then	then	ADV
ijassa-348	299	18	semble	semble	ADJ
ijassa-348	299	19	approach	approach	NOUN
ijassa-348	299	20	overcomes	overcome	VERB
ijassa-348	299	21	the	the	DET
ijassa-348	299	22	difference	difference	NOUN
ijassa-348	299	23	in	in	ADP
ijassa-348	299	24	misclassification	misclassification	NOUN
ijassa-348	299	25	.	.	PUNCT
ijassa-348	300	1	the	the	DET
ijassa-348	300	2	overall	overall	ADJ
ijassa-348	300	3	complexity	complexity	NOUN
ijassa-348	300	4	of	of	ADP
ijassa-348	300	5	the	the	DET
ijassa-348	300	6	model	model	NOUN
ijassa-348	300	7	is	be	AUX
ijassa-348	300	8	o(n4	o(n4	NOUN
ijassa-348	300	9	*	*	PUNCT
ijassa-348	300	10	m	m	NOUN
ijassa-348	300	11	)	)	PUNCT
ijassa-348	300	12	where	where	SCONJ
ijassa-348	300	13	n	n	PRON
ijassa-348	300	14	is	be	AUX
ijassa-348	300	15	the	the	DET
ijassa-348	300	16	number	number	NOUN
ijassa-348	300	17	of	of	ADP
ijassa-348	300	18	attributes	attribute	NOUN
ijassa-348	300	19	and	and	CCONJ
ijassa-348	300	20	m	m	NOUN
ijassa-348	300	21	is	be	AUX
ijassa-348	300	22	the	the	DET
ijassa-348	300	23	number	number	NOUN
ijassa-348	300	24	of	of	ADP
ijassa-348	300	25	training	training	NOUN
ijassa-348	300	26	instances	instance	NOUN
ijassa-348	300	27	.	.	PUNCT
ijassa-348	301	1	ldc	ldc	PROPN
ijassa-348	301	2	performs	perform	VERB
ijassa-348	301	3	computation	computation	NOUN
ijassa-348	301	4	in	in	ADP
ijassa-348	301	5	o(n	o(n	PROPN
ijassa-348	301	6	)	)	PUNCT
ijassa-348	301	7	,	,	PUNCT
ijassa-348	301	8	qdc	qdc	VERB
ijassa-348	301	9	performs	perform	VERB
ijassa-348	301	10	computation	computation	NOUN
ijassa-348	301	11	in	in	ADP
ijassa-348	301	12	o(n2	o(n2	ADV
ijassa-348	301	13	)	)	PUNCT
ijassa-348	301	14	and	and	CCONJ
ijassa-348	301	15	svm	svm	PROPN
ijassa-348	301	16	performs	perform	VERB
ijassa-348	301	17	computation	computation	NOUN
ijassa-348	301	18	in	in	ADP
ijassa-348	301	19	o(nm	o(nm	NOUN
ijassa-348	301	20	)	)	PUNCT
ijassa-348	301	21	time	time	NOUN
ijassa-348	301	22	.	.	PUNCT
ijassa-348	302	1	fig	fig	NOUN
ijassa-348	302	2	.	.	PUNCT
ijassa-348	303	1	1	1	NUM
ijassa-348	303	2	proposed	propose	VERB
ijassa-348	303	3	intrusion	intrusion	NOUN
ijassa-348	303	4	detection	detection	NOUN
ijassa-348	303	5	model	model	NOUN
ijassa-348	303	6	advances	advance	NOUN
ijassa-348	303	7	in	in	ADP
ijassa-348	303	8	systems	system	NOUN
ijassa-348	303	9	science	science	NOUN
ijassa-348	303	10	and	and	CCONJ
ijassa-348	303	11	application	application	NOUN
ijassa-348	303	12	(	(	PUNCT
ijassa-348	303	13	2016	2016	NUM
ijassa-348	303	14	)	)	PUNCT
ijassa-348	304	1	vol.16	vol.16	PROPN
ijassa-348	304	2	no.2	no.2	PROPN
ijassa-348	304	3	29	29	NUM
ijassa-348	304	4	5	5	NUM
ijassa-348	304	5	experiment	experiment	NOUN
ijassa-348	304	6	setup	setup	NOUN
ijassa-348	304	7	and	and	CCONJ
ijassa-348	304	8	results	result	VERB
ijassa-348	304	9	the	the	DET
ijassa-348	304	10	experimental	experimental	ADJ
ijassa-348	304	11	study	study	NOUN
ijassa-348	304	12	was	be	AUX
ijassa-348	304	13	conducted	conduct	VERB
ijassa-348	304	14	with	with	ADP
ijassa-348	304	15	nsl	nsl	NOUN
ijassa-348	304	16	-	-	PUNCT
ijassa-348	304	17	kdd	kdd	PROPN
ijassa-348	304	18	and	and	CCONJ
ijassa-348	304	19	unsw	unsw	PROPN
ijassa-348	304	20	-	-	PUNCT
ijassa-348	304	21	nb	nb	NOUN
ijassa-348	304	22	datasets	dataset	NOUN
ijassa-348	304	23	as	as	SCONJ
ijassa-348	304	24	discussed	discuss	VERB
ijassa-348	304	25	in	in	ADP
ijassa-348	304	26	the	the	DET
ijassa-348	304	27	section	section	NOUN
ijassa-348	304	28	3.2	3.2	NUM
ijassa-348	304	29	.	.	PUNCT
ijassa-348	305	1	the	the	DET
ijassa-348	305	2	experiments	experiment	NOUN
ijassa-348	305	3	utilized	utilize	VERB
ijassa-348	305	4	matlab-2013b	matlab-2013b	NOUN
ijassa-348	305	5	installed	instal	VERB
ijassa-348	305	6	on	on	ADP
ijassa-348	305	7	windows	window	NOUN
ijassa-348	305	8	7	7	NUM
ijassa-348	305	9	ultimate	ultimate	ADJ
ijassa-348	305	10	64	64	NUM
ijassa-348	305	11	-	-	PUNCT
ijassa-348	305	12	bit	bit	NOUN
ijassa-348	305	13	machine	machine	NOUN
ijassa-348	305	14	.	.	PUNCT
ijassa-348	306	1	a	a	DET
ijassa-348	306	2	5	5	NUM
ijassa-348	306	3	-	-	ADJ
ijassa-348	306	4	fold	fold	ADJ
ijassa-348	306	5	cross	cross	NOUN
ijassa-348	306	6	validation	validation	NOUN
ijassa-348	306	7	is	be	AUX
ijassa-348	306	8	performed	perform	VERB
ijassa-348	306	9	on	on	ADP
ijassa-348	306	10	all	all	DET
ijassa-348	306	11	the	the	DET
ijassa-348	306	12	classifiers	classifier	NOUN
ijassa-348	306	13	for	for	ADP
ijassa-348	306	14	splitting	split	VERB
ijassa-348	306	15	the	the	DET
ijassa-348	306	16	dataset	dataset	NOUN
ijassa-348	306	17	into	into	ADP
ijassa-348	306	18	5	5	NUM
ijassa-348	306	19	non	non	NOUN
ijassa-348	306	20	repeated	repeat	VERB
ijassa-348	306	21	subsets	subset	NOUN
ijassa-348	306	22	for	for	ADP
ijassa-348	306	23	training	training	NOUN
ijassa-348	306	24	,	,	PUNCT
ijassa-348	306	25	validation	validation	NOUN
ijassa-348	306	26	and	and	CCONJ
ijassa-348	306	27	testing	testing	NOUN
ijassa-348	306	28	.	.	PUNCT
ijassa-348	307	1	in	in	ADP
ijassa-348	307	2	this	this	DET
ijassa-348	307	3	paper	paper	NOUN
ijassa-348	307	4	we	we	PRON
ijassa-348	307	5	compare	compare	VERB
ijassa-348	307	6	the	the	DET
ijassa-348	307	7	efficiency	efficiency	NOUN
ijassa-348	307	8	of	of	ADP
ijassa-348	307	9	the	the	DET
ijassa-348	307	10	proposed	propose	VERB
ijassa-348	307	11	algorithms	algorithm	NOUN
ijassa-348	307	12	with	with	ADP
ijassa-348	307	13	the	the	DET
ijassa-348	307	14	metrics	metric	NOUN
ijassa-348	307	15	namely	namely	ADV
ijassa-348	307	16	accuracy	accuracy	NOUN
ijassa-348	307	17	rate	rate	NOUN
ijassa-348	307	18	and	and	CCONJ
ijassa-348	307	19	elapsed	elapse	VERB
ijassa-348	307	20	time	time	NOUN
ijassa-348	307	21	.	.	PUNCT
ijassa-348	308	1	•	•	NUM
ijassa-348	308	2	accuracy	accuracy	NOUN
ijassa-348	308	3	:	:	PUNCT
ijassa-348	308	4	number	number	NOUN
ijassa-348	308	5	of	of	ADP
ijassa-348	308	6	test	test	NOUN
ijassa-348	308	7	instances	instance	NOUN
ijassa-348	308	8	correctly	correctly	ADV
ijassa-348	308	9	classified	classify	VERB
ijassa-348	308	10	by	by	ADP
ijassa-348	308	11	the	the	DET
ijassa-348	308	12	model	model	NOUN
ijassa-348	308	13	.	.	PUNCT
ijassa-348	309	1	•	•	NUM
ijassa-348	309	2	elapsed	elapse	VERB
ijassa-348	309	3	time	time	NOUN
ijassa-348	309	4	:	:	PUNCT
ijassa-348	309	5	time	time	NOUN
ijassa-348	309	6	to	to	PART
ijassa-348	309	7	complete	complete	VERB
ijassa-348	309	8	the	the	DET
ijassa-348	309	9	detection	detection	NOUN
ijassa-348	309	10	by	by	ADP
ijassa-348	309	11	a	a	DET
ijassa-348	309	12	classification	classification	NOUN
ijassa-348	309	13	approach	approach	NOUN
ijassa-348	309	14	.	.	PUNCT
ijassa-348	310	1	an	an	DET
ijassa-348	310	2	important	important	ADJ
ijassa-348	310	3	advantage	advantage	NOUN
ijassa-348	310	4	of	of	ADP
ijassa-348	310	5	intergrating	intergrate	VERB
ijassa-348	310	6	complementary	complementary	ADJ
ijassa-348	310	7	classifiers	classifier	NOUN
ijassa-348	310	8	is	be	AUX
ijassa-348	310	9	to	to	PART
ijassa-348	310	10	improve	improve	VERB
ijassa-348	310	11	accuracy	accuracy	NOUN
ijassa-348	310	12	and	and	CCONJ
ijassa-348	310	13	generalization	generalization	NOUN
ijassa-348	310	14	performance	performance	NOUN
ijassa-348	310	15	.	.	PUNCT
ijassa-348	311	1	we	we	PRON
ijassa-348	311	2	assume	assume	VERB
ijassa-348	311	3	the	the	DET
ijassa-348	311	4	accuracy	accuracy	NOUN
ijassa-348	311	5	of	of	ADP
ijassa-348	311	6	each	each	DET
ijassa-348	311	7	classifier	classifier	NOUN
ijassa-348	311	8	separately	separately	ADV
ijassa-348	311	9	.	.	PUNCT
ijassa-348	312	1	the	the	DET
ijassa-348	312	2	ensemble	ensemble	ADJ
ijassa-348	312	3	approach	approach	NOUN
ijassa-348	312	4	is	be	AUX
ijassa-348	312	5	compared	compare	VERB
ijassa-348	312	6	by	by	ADP
ijassa-348	312	7	considering	consider	VERB
ijassa-348	312	8	the	the	DET
ijassa-348	312	9	average	average	ADJ
ijassa-348	312	10	score	score	NOUN
ijassa-348	312	11	from	from	ADP
ijassa-348	312	12	each	each	DET
ijassa-348	312	13	classifier	classifier	NOUN
ijassa-348	312	14	.	.	PUNCT
ijassa-348	313	1	we	we	PRON
ijassa-348	313	2	analyzed	analyze	VERB
ijassa-348	313	3	the	the	DET
ijassa-348	313	4	efficiency	efficiency	NOUN
ijassa-348	313	5	of	of	ADP
ijassa-348	313	6	each	each	DET
ijassa-348	313	7	classifier	classifier	NOUN
ijassa-348	313	8	in	in	ADP
ijassa-348	313	9	the	the	DET
ijassa-348	313	10	base	base	NOUN
ijassa-348	313	11	and	and	CCONJ
ijassa-348	313	12	ensemble	ensemble	ADJ
ijassa-348	313	13	.	.	PUNCT
ijassa-348	314	1	thus	thus	ADV
ijassa-348	314	2	the	the	DET
ijassa-348	314	3	managers	manager	NOUN
ijassa-348	314	4	accuracy	accuracy	NOUN
ijassa-348	314	5	is	be	AUX
ijassa-348	314	6	defined	define	VERB
ijassa-348	314	7	by	by	ADP
ijassa-348	314	8	ei	ei	NOUN
ijassa-348	314	9	=	=	PUNCT
ijassa-348	314	10	∑n	∑n	PROPN
ijassa-348	314	11	i=1ai	i=1ai	ADV
ijassa-348	314	12	n	n	CCONJ
ijassa-348	314	13	(	(	PUNCT
ijassa-348	314	14	20	20	NUM
ijassa-348	314	15	)	)	PUNCT
ijassa-348	314	16	where	where	SCONJ
ijassa-348	314	17	‘	'	PUNCT
ijassa-348	314	18	n	n	CCONJ
ijassa-348	314	19	’	'	PUNCT
ijassa-348	314	20	is	be	AUX
ijassa-348	314	21	the	the	DET
ijassa-348	314	22	number	number	NOUN
ijassa-348	314	23	of	of	ADP
ijassa-348	314	24	classes	class	NOUN
ijassa-348	314	25	.	.	PUNCT
ijassa-348	315	1	5.1	5.1	NUM
ijassa-348	315	2	study	study	NOUN
ijassa-348	315	3	1	1	NUM
ijassa-348	315	4	:	:	PUNCT
ijassa-348	315	5	nsl	nsl	PROPN
ijassa-348	315	6	-	-	PUNCT
ijassa-348	315	7	kdd	kdd	PROPN
ijassa-348	315	8	dataset	dataset	VERB
ijassa-348	315	9	the	the	DET
ijassa-348	315	10	initial	initial	ADJ
ijassa-348	315	11	step	step	NOUN
ijassa-348	315	12	is	be	AUX
ijassa-348	315	13	to	to	PART
ijassa-348	315	14	perform	perform	VERB
ijassa-348	315	15	preprocessing	preprocesse	VERB
ijassa-348	315	16	on	on	ADP
ijassa-348	315	17	the	the	DET
ijassa-348	315	18	data	datum	NOUN
ijassa-348	315	19	and	and	CCONJ
ijassa-348	315	20	dimensionality	dimensionality	NOUN
ijassa-348	315	21	reduction	reduction	NOUN
ijassa-348	315	22	using	use	VERB
ijassa-348	315	23	pca	pca	PROPN
ijassa-348	315	24	to	to	PART
ijassa-348	315	25	remove	remove	VERB
ijassa-348	315	26	the	the	DET
ijassa-348	315	27	noisy	noisy	ADJ
ijassa-348	315	28	attributes	attribute	NOUN
ijassa-348	315	29	in	in	ADP
ijassa-348	315	30	the	the	DET
ijassa-348	315	31	traffic	traffic	NOUN
ijassa-348	315	32	that	that	PRON
ijassa-348	315	33	do	do	AUX
ijassa-348	315	34	not	not	PART
ijassa-348	315	35	contribute	contribute	VERB
ijassa-348	315	36	for	for	ADP
ijassa-348	315	37	classification	classification	NOUN
ijassa-348	315	38	.	.	PUNCT
ijassa-348	316	1	table	table	NOUN
ijassa-348	316	2	1	1	NUM
ijassa-348	316	3	shows	show	VERB
ijassa-348	316	4	the	the	DET
ijassa-348	316	5	features	feature	NOUN
ijassa-348	316	6	retrieved	retrieve	VERB
ijassa-348	316	7	after	after	ADP
ijassa-348	316	8	dimensionality	dimensionality	NOUN
ijassa-348	316	9	reduction	reduction	NOUN
ijassa-348	316	10	on	on	ADP
ijassa-348	316	11	the	the	DET
ijassa-348	316	12	nsl	nsl	NOUN
ijassa-348	316	13	-	-	PUNCT
ijassa-348	316	14	kdd	kdd	PROPN
ijassa-348	316	15	dataset	dataset	NOUN
ijassa-348	316	16	.	.	PUNCT
ijassa-348	317	1	the	the	DET
ijassa-348	317	2	symbolic	symbolic	ADJ
ijassa-348	317	3	features	feature	NOUN
ijassa-348	317	4	and	and	CCONJ
ijassa-348	317	5	the	the	DET
ijassa-348	317	6	features	feature	NOUN
ijassa-348	317	7	with	with	ADP
ijassa-348	317	8	less	less	ADJ
ijassa-348	317	9	variance	variance	NOUN
ijassa-348	317	10	are	be	AUX
ijassa-348	317	11	removed	remove	VERB
ijassa-348	317	12	from	from	ADP
ijassa-348	317	13	the	the	DET
ijassa-348	317	14	dataset	dataset	NOUN
ijassa-348	317	15	.	.	PUNCT
ijassa-348	318	1	experimental	experimental	ADJ
ijassa-348	318	2	results	result	NOUN
ijassa-348	318	3	for	for	ADP
ijassa-348	318	4	each	each	DET
ijassa-348	318	5	manager	manager	NOUN
ijassa-348	318	6	separately	separately	ADV
ijassa-348	318	7	for	for	ADP
ijassa-348	318	8	svm	svm	PROPN
ijassa-348	318	9	,	,	PUNCT
ijassa-348	318	10	ldc	ldc	PROPN
ijassa-348	318	11	,	,	PUNCT
ijassa-348	318	12	qdc	qdc	X
ijassa-348	318	13	and	and	CCONJ
ijassa-348	318	14	ensemble	ensemble	ADJ
ijassa-348	318	15	managers	manager	NOUN
ijassa-348	318	16	on	on	ADP
ijassa-348	318	17	the	the	DET
ijassa-348	318	18	nslkdd	nslkdd	ADJ
ijassa-348	318	19	datasets	dataset	NOUN
ijassa-348	318	20	are	be	AUX
ijassa-348	318	21	specified	specify	VERB
ijassa-348	318	22	in	in	ADP
ijassa-348	318	23	tables	table	NOUN
ijassa-348	318	24	2	2	NUM
ijassa-348	318	25	-	-	SYM
ijassa-348	318	26	5	5	NUM
ijassa-348	318	27	respectively	respectively	ADV
ijassa-348	318	28	.	.	PUNCT
ijassa-348	319	1	the	the	DET
ijassa-348	319	2	results	result	NOUN
ijassa-348	319	3	show	show	VERB
ijassa-348	319	4	that	that	SCONJ
ijassa-348	319	5	by	by	ADP
ijassa-348	319	6	deploying	deploy	VERB
ijassa-348	319	7	different	different	ADJ
ijassa-348	319	8	classifiers	classifier	NOUN
ijassa-348	319	9	for	for	ADP
ijassa-348	319	10	each	each	DET
ijassa-348	319	11	class	class	NOUN
ijassa-348	319	12	,	,	PUNCT
ijassa-348	319	13	higher	high	ADJ
ijassa-348	319	14	accuracy	accuracy	NOUN
ijassa-348	319	15	is	be	AUX
ijassa-348	319	16	achieved	achieve	VERB
ijassa-348	319	17	for	for	ADP
ijassa-348	319	18	all	all	DET
ijassa-348	319	19	classes	class	NOUN
ijassa-348	319	20	namely	namely	ADV
ijassa-348	319	21	normal	normal	ADJ
ijassa-348	319	22	,	,	PUNCT
ijassa-348	319	23	dos	do	NOUN
ijassa-348	319	24	,	,	PUNCT
ijassa-348	319	25	probe	probe	NOUN
ijassa-348	319	26	,	,	PUNCT
ijassa-348	319	27	u2r	u2r	NOUN
ijassa-348	319	28	and	and	CCONJ
ijassa-348	319	29	r2l	r2l	NOUN
ijassa-348	319	30	which	which	PRON
ijassa-348	319	31	is	be	AUX
ijassa-348	319	32	highlighted	highlight	VERB
ijassa-348	319	33	in	in	ADP
ijassa-348	319	34	each	each	DET
ijassa-348	319	35	table	table	NOUN
ijassa-348	319	36	.	.	PUNCT
ijassa-348	320	1	it	it	PRON
ijassa-348	320	2	is	be	AUX
ijassa-348	320	3	also	also	ADV
ijassa-348	320	4	shown	show	VERB
ijassa-348	320	5	that	that	SCONJ
ijassa-348	320	6	in	in	ADP
ijassa-348	320	7	comparison	comparison	NOUN
ijassa-348	320	8	to	to	ADP
ijassa-348	320	9	base	base	NOUN
ijassa-348	320	10	classifiers	classifier	NOUN
ijassa-348	320	11	and	and	CCONJ
ijassa-348	320	12	ensemble	ensemble	ADJ
ijassa-348	320	13	managers	manager	NOUN
ijassa-348	320	14	,	,	PUNCT
ijassa-348	320	15	the	the	DET
ijassa-348	320	16	latter	latter	ADJ
ijassa-348	320	17	results	result	NOUN
ijassa-348	320	18	in	in	ADP
ijassa-348	320	19	high	high	ADJ
ijassa-348	320	20	accuracy	accuracy	NOUN
ijassa-348	320	21	above	above	ADP
ijassa-348	320	22	99	99	NUM
ijassa-348	320	23	%	%	NOUN
ijassa-348	320	24	for	for	ADP
ijassa-348	320	25	all	all	DET
ijassa-348	320	26	the	the	DET
ijassa-348	320	27	class	class	NOUN
ijassa-348	320	28	labels	label	NOUN
ijassa-348	320	29	.	.	PUNCT
ijassa-348	321	1	time	time	NOUN
ijassa-348	321	2	required	require	VERB
ijassa-348	321	3	for	for	ADP
ijassa-348	321	4	classification	classification	NOUN
ijassa-348	321	5	for	for	ADP
ijassa-348	321	6	each	each	PRON
ijassa-348	321	7	of	of	ADP
ijassa-348	321	8	the	the	DET
ijassa-348	321	9	managers	manager	NOUN
ijassa-348	321	10	is	be	AUX
ijassa-348	321	11	presented	present	VERB
ijassa-348	321	12	in	in	ADP
ijassa-348	321	13	table	table	NOUN
ijassa-348	321	14	6	6	NUM
ijassa-348	321	15	.	.	PUNCT
ijassa-348	322	1	the	the	DET
ijassa-348	322	2	time	time	NOUN
ijassa-348	322	3	consumption	consumption	NOUN
ijassa-348	322	4	for	for	ADP
ijassa-348	322	5	the	the	DET
ijassa-348	322	6	ensemble	ensemble	ADJ
ijassa-348	322	7	wma	wma	NOUN
ijassa-348	322	8	is	be	AUX
ijassa-348	322	9	relatively	relatively	ADV
ijassa-348	322	10	higher	high	ADJ
ijassa-348	322	11	in	in	ADP
ijassa-348	322	12	comparison	comparison	NOUN
ijassa-348	322	13	to	to	ADP
ijassa-348	322	14	base	base	NOUN
ijassa-348	322	15	classifiers	classifier	NOUN
ijassa-348	322	16	but	but	CCONJ
ijassa-348	322	17	it	it	PRON
ijassa-348	322	18	can	can	AUX
ijassa-348	322	19	be	be	AUX
ijassa-348	322	20	neglected	neglect	VERB
ijassa-348	322	21	as	as	SCONJ
ijassa-348	322	22	the	the	DET
ijassa-348	322	23	accuracy	accuracy	NOUN
ijassa-348	322	24	is	be	AUX
ijassa-348	322	25	high	high	ADJ
ijassa-348	322	26	.	.	PUNCT
ijassa-348	323	1	figs	fig	NOUN
ijassa-348	323	2	2	2	NUM
ijassa-348	323	3	-	-	SYM
ijassa-348	323	4	5	5	NUM
ijassa-348	323	5	represent	represent	VERB
ijassa-348	323	6	the	the	DET
ijassa-348	323	7	accuracies	accuracy	NOUN
ijassa-348	323	8	of	of	ADP
ijassa-348	323	9	svm	svm	PROPN
ijassa-348	323	10	,	,	PUNCT
ijassa-348	323	11	ldc	ldc	PROPN
ijassa-348	323	12	,	,	PUNCT
ijassa-348	323	13	qdc	qdc	PROPN
ijassa-348	323	14	and	and	CCONJ
ijassa-348	323	15	ensemble	ensemble	ADJ
ijassa-348	323	16	wma	wma	NOUN
ijassa-348	323	17	managers	manager	NOUN
ijassa-348	323	18	respectively	respectively	ADV
ijassa-348	323	19	.	.	PUNCT
ijassa-348	324	1	it	it	PRON
ijassa-348	324	2	is	be	AUX
ijassa-348	324	3	evident	evident	ADJ
ijassa-348	324	4	that	that	SCONJ
ijassa-348	324	5	wma	wma	NOUN
ijassa-348	324	6	produces	produce	VERB
ijassa-348	324	7	highest	high	ADJ
ijassa-348	324	8	accuracy	accuracy	NOUN
ijassa-348	324	9	in	in	ADP
ijassa-348	324	10	comparison	comparison	NOUN
ijassa-348	324	11	to	to	ADP
ijassa-348	324	12	all	all	DET
ijassa-348	324	13	base	base	NOUN
ijassa-348	324	14	classifiers	classifier	NOUN
ijassa-348	324	15	though	though	SCONJ
ijassa-348	324	16	individually	individually	ADV
ijassa-348	324	17	svm	svm	PROPN
ijassa-348	324	18	,	,	PUNCT
ijassa-348	324	19	ldc	ldc	PROPN
ijassa-348	324	20	and	and	CCONJ
ijassa-348	324	21	qdc	qdc	PROPN
ijassa-348	324	22	produces	produce	VERB
ijassa-348	324	23	optimal	optimal	ADJ
ijassa-348	324	24	results	result	NOUN
ijassa-348	324	25	for	for	ADP
ijassa-348	324	26	a	a	DET
ijassa-348	324	27	single	single	ADJ
ijassa-348	324	28	or	or	CCONJ
ijassa-348	324	29	two	two	NUM
ijassa-348	324	30	class	class	NOUN
ijassa-348	324	31	label	label	NOUN
ijassa-348	324	32	only	only	ADV
ijassa-348	324	33	.	.	PUNCT
ijassa-348	325	1	30	30	NUM
ijassa-348	325	2	sumaiya	sumaiya	NOUN
ijassa-348	325	3	thaseen	thaseen	NOUN
ijassa-348	325	4	and	and	CCONJ
ijassa-348	325	5	ch.aswani	ch.aswani	X
ijassa-348	326	1	kumar	kumar	PROPN
ijassa-348	326	2	:	:	PUNCT
ijassa-348	326	3	intrusion	intrusion	NOUN
ijassa-348	326	4	detection	detection	NOUN
ijassa-348	326	5	model	model	NOUN
ijassa-348	326	6	using	use	VERB
ijassa-348	326	7	pca	pca	PROPN
ijassa-348	326	8	and	and	CCONJ
ijassa-348	326	9	...	...	PUNCT
ijassa-348	326	10	table	table	NOUN
ijassa-348	326	11	1	1	NUM
ijassa-348	326	12	features	feature	NOUN
ijassa-348	326	13	selected	select	VERB
ijassa-348	326	14	by	by	ADP
ijassa-348	326	15	pca	pca	PROPN
ijassa-348	326	16	in	in	ADP
ijassa-348	326	17	nsl	nsl	PROPN
ijassa-348	326	18	-	-	PUNCT
ijassa-348	326	19	kdd	kdd	PROPN
ijassa-348	326	20	dataset	dataset	NOUN
ijassa-348	326	21	22	22	NUM
ijassa-348	326	22	features	feature	NOUN
ijassa-348	326	23	selected	select	VERB
ijassa-348	326	24	after	after	ADP
ijassa-348	326	25	dimensionality	dimensionality	NOUN
ijassa-348	326	26	reduction	reduction	NOUN
ijassa-348	326	27	in	in	ADP
ijassa-348	326	28	nslkdd	nslkdd	ADJ
ijassa-348	326	29	data	datum	NOUN
ijassa-348	326	30	set	set	VERB
ijassa-348	326	31	.	.	PUNCT
ijassa-348	327	1	service	service	NOUN
ijassa-348	327	2	,	,	PUNCT
ijassa-348	327	3	dst	dst	NOUN
ijassa-348	327	4	bytes	byte	NOUN
ijassa-348	327	5	,	,	PUNCT
ijassa-348	327	6	dst	dst	PROPN
ijassa-348	327	7	host	host	NOUN
ijassa-348	327	8	diff	diff	PROPN
ijassa-348	327	9	srv	srv	PROPN
ijassa-348	327	10	rate	rate	NOUN
ijassa-348	327	11	,	,	PUNCT
ijassa-348	327	12	flag	flag	NOUN
ijassa-348	327	13	,	,	PUNCT
ijassa-348	327	14	dst	dst	NOUN
ijassa-348	327	15	host	host	NOUN
ijassa-348	327	16	serror	serror	NOUN
ijassa-348	327	17	rate	rate	NOUN
ijassa-348	327	18	,	,	PUNCT
ijassa-348	327	19	dst	dst	PROPN
ijassa-348	327	20	host	host	NOUN
ijassa-348	327	21	srv	srv	PROPN
ijassa-348	327	22	count	count	PROPN
ijassa-348	327	23	,	,	PUNCT
ijassa-348	327	24	same	same	ADJ
ijassa-348	327	25	srv	srv	PROPN
ijassa-348	327	26	rate	rate	NOUN
ijassa-348	327	27	,	,	PUNCT
ijassa-348	327	28	dst	dst	PROPN
ijassa-348	327	29	host	host	PROPN
ijassa-348	327	30	same	same	ADJ
ijassa-348	327	31	srv	srv	PROPN
ijassa-348	327	32	rate	rate	NOUN
ijassa-348	327	33	,	,	PUNCT
ijassa-348	327	34	serror	serror	NOUN
ijassa-348	327	35	rate	rate	NOUN
ijassa-348	327	36	,	,	PUNCT
ijassa-348	327	37	src	src	NOUN
ijassa-348	327	38	bytes	byte	NOUN
ijassa-348	327	39	,	,	PUNCT
ijassa-348	327	40	dst	dst	PROPN
ijassa-348	327	41	host	host	NOUN
ijassa-348	327	42	srv	srv	PROPN
ijassa-348	327	43	diff	diff	PROPN
ijassa-348	327	44	host	host	NOUN
ijassa-348	327	45	rate	rate	NOUN
ijassa-348	327	46	,	,	PUNCT
ijassa-348	327	47	host	host	NOUN
ijassa-348	327	48	,	,	PUNCT
ijassa-348	327	49	dst	dst	PROPN
ijassa-348	327	50	host	host	NOUN
ijassa-348	327	51	rerror	rerror	PROPN
ijassa-348	327	52	rate	rate	NOUN
ijassa-348	327	53	duration	duration	NOUN
ijassa-348	327	54	,	,	PUNCT
ijassa-348	327	55	srv	srv	PROPN
ijassa-348	327	56	diff	diff	PROPN
ijassa-348	327	57	host	host	NOUN
ijassa-348	327	58	rate	rate	NOUN
ijassa-348	327	59	,	,	PUNCT
ijassa-348	327	60	dst	dst	PROPN
ijassa-348	327	61	host	host	NOUN
ijassa-348	327	62	srv	srv	PROPN
ijassa-348	327	63	rerror	rerror	PROPN
ijassa-348	327	64	rate	rate	NOUN
ijassa-348	327	65	er	er	INTJ
ijassa-348	327	66	ror	ror	NOUN
ijassa-348	327	67	rateprotocol	rateprotocol	NOUN
ijassa-348	327	68	type	type	NOUN
ijassa-348	327	69	,	,	PUNCT
ijassa-348	327	70	srv	srv	PROPN
ijassa-348	327	71	rerror	rerror	NOUN
ijassa-348	327	72	rate	rate	NOUN
ijassa-348	327	73	,	,	PUNCT
ijassa-348	327	74	is	be	AUX
ijassa-348	327	75	guest	guest	NOUN
ijassa-348	327	76	login	login	NOUN
ijassa-348	327	77	,	,	PUNCT
ijassa-348	327	78	srv	srv	PROPN
ijassa-348	327	79	count	count	NOUN
ijassa-348	327	80	,	,	PUNCT
ijassa-348	327	81	num	num	PRON
ijassa-348	327	82	compromised	compromise	VERB
ijassa-348	327	83	.	.	PUNCT
ijassa-348	328	1	table	table	NOUN
ijassa-348	328	2	2	2	NUM
ijassa-348	328	3	accuracy	accuracy	NOUN
ijassa-348	328	4	obtained	obtain	VERB
ijassa-348	328	5	using	use	VERB
ijassa-348	328	6	svm	svm	ADJ
ijassa-348	328	7	manager	manager	NOUN
ijassa-348	328	8	with	with	ADP
ijassa-348	328	9	different	different	ADJ
ijassa-348	328	10	rbf	rbf	PROPN
ijassa-348	328	11	kernel	kernel	PROPN
ijassa-348	328	12	in	in	ADP
ijassa-348	328	13	nsl	nsl	PROPN
ijassa-348	328	14	-	-	PUNCT
ijassa-348	328	15	kdd	kdd	PROPN
ijassa-348	328	16	dataset	dataset	NOUN
ijassa-348	328	17	manager	manager	NOUN
ijassa-348	328	18	normal	normal	ADJ
ijassa-348	328	19	probe	probe	NOUN
ijassa-348	328	20	dos	do	NOUN
ijassa-348	328	21	u2r	u2r	PROPN
ijassa-348	328	22	r2l	r2l	NOUN
ijassa-348	328	23	svm	svm	VERB
ijassa-348	328	24	1	1	NUM
ijassa-348	328	25	(	(	PUNCT
ijassa-348	328	26	rbf:5	rbf:5	NOUN
ijassa-348	328	27	)	)	PUNCT
ijassa-348	328	28	92.9	92.9	NUM
ijassa-348	328	29	98.08	98.08	NUM
ijassa-348	328	30	74.41	74.41	NUM
ijassa-348	328	31	90.08	90.08	NUM
ijassa-348	328	32	87.5	87.5	NUM
ijassa-348	328	33	svm	svm	NOUN
ijassa-348	328	34	2	2	NUM
ijassa-348	328	35	(	(	PUNCT
ijassa-348	328	36	rbf:2	rbf:2	X
ijassa-348	328	37	)	)	PUNCT
ijassa-348	328	38	91.93	91.93	NUM
ijassa-348	328	39	97.79	97.79	NUM
ijassa-348	328	40	51.42	51.42	NUM
ijassa-348	328	41	84.1	84.1	NUM
ijassa-348	328	42	100	100	NUM
ijassa-348	328	43	svm	svm	NOUN
ijassa-348	328	44	3	3	NUM
ijassa-348	328	45	(	(	PUNCT
ijassa-348	328	46	rbf	rbf	PROPN
ijassa-348	328	47	:	:	PUNCT
ijassa-348	328	48	1	1	NUM
ijassa-348	328	49	)	)	PUNCT
ijassa-348	328	50	90.39	90.39	NUM
ijassa-348	328	51	99.8	99.8	NUM
ijassa-348	328	52	85.71	85.71	NUM
ijassa-348	328	53	85.2	85.2	NUM
ijassa-348	328	54	88.88	88.88	NUM
ijassa-348	328	55	svm	svm	NOUN
ijassa-348	328	56	4	4	NUM
ijassa-348	328	57	(	(	PUNCT
ijassa-348	328	58	rbf:0.5	rbf:0.5	NOUN
ijassa-348	328	59	)	)	PUNCT
ijassa-348	328	60	89.73	89.73	NUM
ijassa-348	328	61	97.78	97.78	NUM
ijassa-348	328	62	71.82	71.82	NUM
ijassa-348	328	63	100	100	NUM
ijassa-348	328	64	83.3	83.3	NUM
ijassa-348	328	65	svm5	svm5	NOUN
ijassa-348	328	66	(	(	PUNCT
ijassa-348	328	67	rbf:0.2	rbf:0.2	NOUN
ijassa-348	328	68	)	)	PUNCT
ijassa-348	328	69	99.54	99.54	NUM
ijassa-348	328	70	97.94	97.94	NUM
ijassa-348	328	71	56.25	56.25	NUM
ijassa-348	328	72	99.2	99.2	NUM
ijassa-348	328	73	89.47	89.47	NUM
ijassa-348	328	74	table	table	NOUN
ijassa-348	328	75	3	3	NUM
ijassa-348	328	76	accuracy	accuracy	NOUN
ijassa-348	328	77	obtained	obtain	VERB
ijassa-348	328	78	by	by	ADP
ijassa-348	328	79	linear	linear	PROPN
ijassa-348	328	80	discriminant	discriminant	ADJ
ijassa-348	328	81	analysis	analysis	NOUN
ijassa-348	328	82	in	in	ADP
ijassa-348	328	83	nsl	nsl	NOUN
ijassa-348	328	84	-	-	PUNCT
ijassa-348	328	85	kdd	kdd	PROPN
ijassa-348	328	86	dataset	dataset	NOUN
ijassa-348	328	87	manager	manager	NOUN
ijassa-348	328	88	normal	normal	ADJ
ijassa-348	328	89	dos	do	NOUN
ijassa-348	328	90	probe	probe	VERB
ijassa-348	328	91	u2r	u2r	PROPN
ijassa-348	328	92	r2l	r2l	X
ijassa-348	328	93	ldc	ldc	PROPN
ijassa-348	328	94	1	1	NUM
ijassa-348	328	95	98.87	98.87	NUM
ijassa-348	328	96	99.83	99.83	NUM
ijassa-348	328	97	82.05	82.05	NUM
ijassa-348	328	98	96.99	96.99	NUM
ijassa-348	328	99	99.12	99.12	NUM
ijassa-348	328	100	ldc	ldc	PROPN
ijassa-348	328	101	2	2	NUM
ijassa-348	328	102	93.22	93.22	NUM
ijassa-348	328	103	97.79	97.79	NUM
ijassa-348	328	104	94.12	94.12	NUM
ijassa-348	328	105	86.77	86.77	NUM
ijassa-348	328	106	99.65	99.65	NUM
ijassa-348	328	107	ldc	ldc	PROPN
ijassa-348	328	108	3	3	NUM
ijassa-348	328	109	96.58	96.58	NUM
ijassa-348	328	110	98.56	98.56	NUM
ijassa-348	328	111	100	100	NUM
ijassa-348	328	112	99.45	99.45	NUM
ijassa-348	328	113	98.56	98.56	NUM
ijassa-348	328	114	ldc	ldc	PROPN
ijassa-348	328	115	4	4	NUM
ijassa-348	328	116	99.58	99.58	NUM
ijassa-348	328	117	99.6	99.6	NUM
ijassa-348	328	118	86.48	86.48	NUM
ijassa-348	328	119	100	100	NUM
ijassa-348	328	120	81.81	81.81	NUM
ijassa-348	328	121	ldc	ldc	PROPN
ijassa-348	328	122	5	5	NUM
ijassa-348	328	123	70.4	70.4	NUM
ijassa-348	328	124	93.92	93.92	NUM
ijassa-348	328	125	71.84	71.84	NUM
ijassa-348	328	126	55.86	55.86	NUM
ijassa-348	328	127	97.42	97.42	NUM
ijassa-348	328	128	table	table	NOUN
ijassa-348	328	129	4	4	NUM
ijassa-348	328	130	accuracy	accuracy	NOUN
ijassa-348	328	131	obtained	obtain	VERB
ijassa-348	328	132	by	by	ADP
ijassa-348	328	133	quadratic	quadratic	ADJ
ijassa-348	328	134	discriminant	discriminant	ADJ
ijassa-348	328	135	analysis	analysis	NOUN
ijassa-348	328	136	in	in	ADP
ijassa-348	328	137	nsl	nsl	NOUN
ijassa-348	328	138	-	-	PUNCT
ijassa-348	328	139	kdd	kdd	PROPN
ijassa-348	328	140	dataset	dataset	NOUN
ijassa-348	328	141	manager	manager	NOUN
ijassa-348	328	142	normal	normal	ADJ
ijassa-348	328	143	probe	probe	NOUN
ijassa-348	328	144	dos	do	NOUN
ijassa-348	328	145	u2r	u2r	NOUN
ijassa-348	328	146	r2l	r2l	X
ijassa-348	328	147	qdc	qdc	VERB
ijassa-348	328	148	1	1	NUM
ijassa-348	328	149	92.898	92.898	NUM
ijassa-348	328	150	97.94	97.94	NUM
ijassa-348	328	151	82.75	82.75	NUM
ijassa-348	328	152	83.81	83.81	NUM
ijassa-348	328	153	89.83	89.83	NUM
ijassa-348	328	154	qdc	qdc	NOUN
ijassa-348	328	155	2	2	NUM
ijassa-348	328	156	91.73	91.73	NUM
ijassa-348	328	157	98.28	98.28	NUM
ijassa-348	328	158	86.29	86.29	NUM
ijassa-348	328	159	86.81	86.81	NUM
ijassa-348	328	160	93.32	93.32	NUM
ijassa-348	328	161	qdc	qdc	NOUN
ijassa-348	328	162	3	3	NUM
ijassa-348	328	163	87.17	87.17	NUM
ijassa-348	328	164	97.55	97.55	NUM
ijassa-348	328	165	67.92	67.92	NUM
ijassa-348	328	166	91.71	91.71	NUM
ijassa-348	328	167	92.21	92.21	NUM
ijassa-348	328	168	qdc	qdc	PROPN
ijassa-348	328	169	4	4	NUM
ijassa-348	328	170	90.59	90.59	NUM
ijassa-348	328	171	97.92	97.92	NUM
ijassa-348	328	172	94.05	94.05	NUM
ijassa-348	328	173	87.44	87.44	NUM
ijassa-348	328	174	92.45	92.45	NUM
ijassa-348	328	175	qdc	qdc	PROPN
ijassa-348	328	176	5	5	NUM
ijassa-348	328	177	89.83	89.83	NUM
ijassa-348	328	178	97.8	97.8	NUM
ijassa-348	328	179	78.98	78.98	NUM
ijassa-348	328	180	86.02	86.02	NUM
ijassa-348	328	181	95.31	95.31	NUM
ijassa-348	328	182	advances	advance	NOUN
ijassa-348	328	183	in	in	ADP
ijassa-348	328	184	systems	system	NOUN
ijassa-348	328	185	science	science	NOUN
ijassa-348	328	186	and	and	CCONJ
ijassa-348	328	187	application	application	NOUN
ijassa-348	328	188	(	(	PUNCT
ijassa-348	328	189	2016	2016	NUM
ijassa-348	328	190	)	)	PUNCT
ijassa-348	328	191	vol.16	vol.16	PROPN
ijassa-348	328	192	no.2	no.2	PROPN
ijassa-348	328	193	31	31	NUM
ijassa-348	328	194	table	table	NOUN
ijassa-348	328	195	5	5	NUM
ijassa-348	328	196	accuracy	accuracy	NOUN
ijassa-348	328	197	obtained	obtain	VERB
ijassa-348	328	198	by	by	ADP
ijassa-348	328	199	wma	wma	NOUN
ijassa-348	328	200	managers	manager	NOUN
ijassa-348	328	201	in	in	ADP
ijassa-348	328	202	nsl	nsl	PROPN
ijassa-348	328	203	-	-	PUNCT
ijassa-348	328	204	kdd	kdd	PROPN
ijassa-348	328	205	dataset	dataset	NOUN
ijassa-348	328	206	manager	manager	NOUN
ijassa-348	328	207	normal	normal	ADJ
ijassa-348	328	208	probe	probe	NOUN
ijassa-348	328	209	dos	do	NOUN
ijassa-348	328	210	u2r	u2r	PROPN
ijassa-348	328	211	r2l	r2l	NOUN
ijassa-348	328	212	wma	wma	VERB
ijassa-348	328	213	1	1	NUM
ijassa-348	328	214	98.93	98.93	NUM
ijassa-348	328	215	99.47	99.47	NUM
ijassa-348	328	216	94.61	94.61	NUM
ijassa-348	328	217	99.45	99.45	NUM
ijassa-348	328	218	99.11	99.11	NUM
ijassa-348	328	219	wma	wma	NOUN
ijassa-348	328	220	2	2	NUM
ijassa-348	328	221	78.77	78.77	NUM
ijassa-348	328	222	94.47	94.47	NUM
ijassa-348	328	223	94.59	94.59	NUM
ijassa-348	328	224	82.78	82.78	NUM
ijassa-348	328	225	99.6	99.6	NUM
ijassa-348	328	226	wma	wma	NOUN
ijassa-348	328	227	3	3	NUM
ijassa-348	328	228	78.89	78.89	NUM
ijassa-348	328	229	94.31	94.31	NUM
ijassa-348	328	230	88.1	88.1	NUM
ijassa-348	328	231	82.36	82.36	NUM
ijassa-348	328	232	84.85	84.85	NUM
ijassa-348	328	233	fig	fig	NOUN
ijassa-348	328	234	.	.	PUNCT
ijassa-348	329	1	2	2	NUM
ijassa-348	329	2	accuracy	accuracy	NOUN
ijassa-348	329	3	of	of	ADP
ijassa-348	329	4	svm	svm	ADJ
ijassa-348	329	5	managers	manager	NOUN
ijassa-348	329	6	in	in	ADP
ijassa-348	329	7	kdd	kdd	PROPN
ijassa-348	329	8	dataset	dataset	PROPN
ijassa-348	329	9	fig	fig	NOUN
ijassa-348	329	10	.	.	PUNCT
ijassa-348	330	1	3	3	NUM
ijassa-348	330	2	accuracy	accuracy	NOUN
ijassa-348	330	3	of	of	ADP
ijassa-348	330	4	ldc	ldc	PROPN
ijassa-348	330	5	managers	manager	NOUN
ijassa-348	330	6	in	in	ADP
ijassa-348	330	7	nsl	nsl	PROPN
ijassa-348	330	8	-	-	PUNCT
ijassa-348	330	9	kdd	kdd	PROPN
ijassa-348	330	10	dataset	dataset	PROPN
ijassa-348	330	11	fig	fig	NOUN
ijassa-348	330	12	.	.	PUNCT
ijassa-348	331	1	4	4	NUM
ijassa-348	331	2	accuracy	accuracy	NOUN
ijassa-348	331	3	of	of	ADP
ijassa-348	331	4	qdc	qdc	PROPN
ijassa-348	331	5	managers	manager	NOUN
ijassa-348	331	6	in	in	ADP
ijassa-348	331	7	nsl	nsl	PROPN
ijassa-348	331	8	-	-	PUNCT
ijassa-348	331	9	kdd	kdd	PROPN
ijassa-348	331	10	dataset	dataset	PROPN
ijassa-348	331	11	fig	fig	NOUN
ijassa-348	331	12	.	.	PUNCT
ijassa-348	332	1	5	5	NUM
ijassa-348	332	2	accuracy	accuracy	NOUN
ijassa-348	332	3	of	of	ADP
ijassa-348	332	4	wma	wma	NOUN
ijassa-348	332	5	managers	manager	NOUN
ijassa-348	332	6	in	in	ADP
ijassa-348	332	7	nsl	nsl	PROPN
ijassa-348	332	8	-	-	PUNCT
ijassa-348	332	9	kdd	kdd	PROPN
ijassa-348	332	10	dataset	dataset	NOUN
ijassa-348	332	11	32	32	NUM
ijassa-348	332	12	sumaiya	sumaiya	NOUN
ijassa-348	332	13	thaseen	thaseen	PROPN
ijassa-348	332	14	and	and	CCONJ
ijassa-348	332	15	ch.aswani	ch.aswani	X
ijassa-348	333	1	kumar	kumar	PROPN
ijassa-348	333	2	:	:	PUNCT
ijassa-348	333	3	intrusion	intrusion	NOUN
ijassa-348	333	4	detection	detection	NOUN
ijassa-348	333	5	model	model	NOUN
ijassa-348	333	6	using	use	VERB
ijassa-348	333	7	pca	pca	PROPN
ijassa-348	333	8	and	and	CCONJ
ijassa-348	333	9	...	...	PUNCT
ijassa-348	333	10	table	table	NOUN
ijassa-348	333	11	6	6	NUM
ijassa-348	333	12	elapsed	elapse	VERB
ijassa-348	333	13	time	time	NOUN
ijassa-348	333	14	for	for	SCONJ
ijassa-348	333	15	managers	manager	NOUN
ijassa-348	333	16	in	in	ADP
ijassa-348	333	17	nsl	nsl	PROPN
ijassa-348	333	18	-	-	PUNCT
ijassa-348	333	19	kdd	kdd	PROPN
ijassa-348	333	20	managers	manager	NOUN
ijassa-348	333	21	elapsed	elapse	VERB
ijassa-348	333	22	time	time	NOUN
ijassa-348	333	23	(	(	PUNCT
ijassa-348	333	24	secs	secs	X
ijassa-348	333	25	)	)	PUNCT
ijassa-348	333	26	svm	svm	VERB
ijassa-348	333	27	31.96	31.96	NUM
ijassa-348	333	28	s	s	PART
ijassa-348	333	29	ldc	ldc	PROPN
ijassa-348	333	30	63.95	63.95	NUM
ijassa-348	333	31	s	s	PART
ijassa-348	333	32	qdc	qdc	NOUN
ijassa-348	333	33	66.09	66.09	NUM
ijassa-348	333	34	s	s	NOUN
ijassa-348	333	35	wma	wma	NOUN
ijassa-348	333	36	53.36	53.36	NUM
ijassa-348	333	37	s	s	PART
ijassa-348	333	38	5.2	5.2	NUM
ijassa-348	333	39	study	study	NOUN
ijassa-348	333	40	ii	ii	NOUN
ijassa-348	333	41	:	:	PUNCT
ijassa-348	333	42	unsw	unsw	PROPN
ijassa-348	333	43	-	-	PUNCT
ijassa-348	333	44	nb	nb	NOUN
ijassa-348	333	45	dataset	dataset	VERB
ijassa-348	333	46	after	after	ADP
ijassa-348	333	47	preprocessing	preprocesse	VERB
ijassa-348	333	48	on	on	ADP
ijassa-348	333	49	the	the	DET
ijassa-348	333	50	dataset	dataset	NOUN
ijassa-348	333	51	,	,	PUNCT
ijassa-348	333	52	dimensionality	dimensionality	NOUN
ijassa-348	333	53	reduction	reduction	NOUN
ijassa-348	333	54	using	use	VERB
ijassa-348	333	55	pca	pca	PROPN
ijassa-348	333	56	is	be	AUX
ijassa-348	333	57	performed	perform	VERB
ijassa-348	333	58	and	and	CCONJ
ijassa-348	333	59	the	the	DET
ijassa-348	333	60	results	result	NOUN
ijassa-348	333	61	are	be	AUX
ijassa-348	333	62	shown	show	VERB
ijassa-348	333	63	in	in	ADP
ijassa-348	333	64	table	table	NOUN
ijassa-348	333	65	7	7	NUM
ijassa-348	333	66	.	.	PUNCT
ijassa-348	334	1	experimental	experimental	ADJ
ijassa-348	334	2	results	result	NOUN
ijassa-348	334	3	for	for	ADP
ijassa-348	334	4	each	each	DET
ijassa-348	334	5	manager	manager	NOUN
ijassa-348	334	6	separately	separately	ADV
ijassa-348	334	7	for	for	ADP
ijassa-348	334	8	svm	svm	PROPN
ijassa-348	334	9	,	,	PUNCT
ijassa-348	334	10	ldc	ldc	PROPN
ijassa-348	334	11	,	,	PUNCT
ijassa-348	334	12	qdc	qdc	X
ijassa-348	334	13	and	and	CCONJ
ijassa-348	334	14	ensemble	ensemble	ADJ
ijassa-348	334	15	manager	manager	NOUN
ijassa-348	334	16	on	on	ADP
ijassa-348	334	17	the	the	DET
ijassa-348	334	18	unsw	unsw	PROPN
ijassa-348	334	19	-	-	PUNCT
ijassa-348	334	20	nb	nb	NOUN
ijassa-348	334	21	datasets	dataset	NOUN
ijassa-348	334	22	are	be	AUX
ijassa-348	334	23	specified	specify	VERB
ijassa-348	334	24	in	in	ADP
ijassa-348	334	25	tables	table	NOUN
ijassa-348	334	26	811	811	NUM
ijassa-348	334	27	respectively	respectively	ADV
ijassa-348	334	28	.	.	PUNCT
ijassa-348	335	1	the	the	DET
ijassa-348	335	2	results	result	NOUN
ijassa-348	335	3	show	show	VERB
ijassa-348	335	4	that	that	SCONJ
ijassa-348	335	5	by	by	ADP
ijassa-348	335	6	deploying	deploy	VERB
ijassa-348	335	7	unique	unique	ADJ
ijassa-348	335	8	classifiers	classifier	NOUN
ijassa-348	335	9	for	for	ADP
ijassa-348	335	10	each	each	DET
ijassa-348	335	11	class	class	NOUN
ijassa-348	335	12	higher	high	ADJ
ijassa-348	335	13	accuracies	accuracy	NOUN
ijassa-348	335	14	are	be	AUX
ijassa-348	335	15	obtained	obtain	VERB
ijassa-348	335	16	for	for	ADP
ijassa-348	335	17	all	all	DET
ijassa-348	335	18	the	the	DET
ijassa-348	335	19	class	class	NOUN
ijassa-348	335	20	labels	label	VERB
ijassa-348	335	21	namely	namely	ADV
ijassa-348	335	22	normal	normal	ADJ
ijassa-348	335	23	,	,	PUNCT
ijassa-348	335	24	analysis	analysis	NOUN
ijassa-348	335	25	,	,	PUNCT
ijassa-348	335	26	backdoor	backdoor	NOUN
ijassa-348	335	27	,	,	PUNCT
ijassa-348	335	28	exploits	exploit	NOUN
ijassa-348	335	29	,	,	PUNCT
ijassa-348	335	30	reconnaissance	reconnaissance	NOUN
ijassa-348	335	31	,	,	PUNCT
ijassa-348	335	32	fuzzers	fuzzer	NOUN
ijassa-348	335	33	,	,	PUNCT
ijassa-348	335	34	exploits	exploit	NOUN
ijassa-348	335	35	,	,	PUNCT
ijassa-348	335	36	dos	do	NOUN
ijassa-348	335	37	and	and	CCONJ
ijassa-348	335	38	shellcode	shellcode	NOUN
ijassa-348	335	39	respectively	respectively	ADV
ijassa-348	335	40	.	.	PUNCT
ijassa-348	336	1	the	the	DET
ijassa-348	336	2	highest	high	ADJ
ijassa-348	336	3	accuracy	accuracy	NOUN
ijassa-348	336	4	for	for	ADP
ijassa-348	336	5	all	all	DET
ijassa-348	336	6	the	the	DET
ijassa-348	336	7	class	class	NOUN
ijassa-348	336	8	labels	label	NOUN
ijassa-348	336	9	are	be	AUX
ijassa-348	336	10	obtained	obtain	VERB
ijassa-348	336	11	using	use	VERB
ijassa-348	336	12	ensemble	ensemble	ADJ
ijassa-348	336	13	manager	manager	NOUN
ijassa-348	336	14	higher	high	ADJ
ijassa-348	336	15	than	than	ADP
ijassa-348	336	16	the	the	DET
ijassa-348	336	17	base	base	NOUN
ijassa-348	336	18	classifiers	classifier	NOUN
ijassa-348	336	19	as	as	SCONJ
ijassa-348	336	20	weighted	weight	VERB
ijassa-348	336	21	majority	majority	NOUN
ijassa-348	336	22	approach	approach	NOUN
ijassa-348	336	23	reduces	reduce	VERB
ijassa-348	336	24	the	the	DET
ijassa-348	336	25	misclassification	misclassification	NOUN
ijassa-348	336	26	rate	rate	NOUN
ijassa-348	336	27	.	.	PUNCT
ijassa-348	337	1	time	time	NOUN
ijassa-348	337	2	required	require	VERB
ijassa-348	337	3	for	for	ADP
ijassa-348	337	4	classification	classification	NOUN
ijassa-348	337	5	for	for	ADP
ijassa-348	337	6	each	each	PRON
ijassa-348	337	7	of	of	ADP
ijassa-348	337	8	the	the	DET
ijassa-348	337	9	managers	manager	NOUN
ijassa-348	337	10	is	be	AUX
ijassa-348	337	11	presented	present	VERB
ijassa-348	337	12	in	in	ADP
ijassa-348	337	13	table	table	NOUN
ijassa-348	337	14	12	12	NUM
ijassa-348	337	15	.	.	PUNCT
ijassa-348	338	1	the	the	DET
ijassa-348	338	2	time	time	NOUN
ijassa-348	338	3	consumption	consumption	NOUN
ijassa-348	338	4	for	for	ADP
ijassa-348	338	5	the	the	DET
ijassa-348	338	6	ensemble	ensemble	ADJ
ijassa-348	338	7	wma	wma	NOUN
ijassa-348	338	8	is	be	AUX
ijassa-348	338	9	relatively	relatively	ADV
ijassa-348	338	10	higher	high	ADJ
ijassa-348	338	11	in	in	ADP
ijassa-348	338	12	comparison	comparison	NOUN
ijassa-348	338	13	to	to	ADP
ijassa-348	338	14	base	base	NOUN
ijassa-348	338	15	classifiers	classifier	NOUN
ijassa-348	338	16	but	but	CCONJ
ijassa-348	338	17	as	as	SCONJ
ijassa-348	338	18	the	the	DET
ijassa-348	338	19	dataset	dataset	NOUN
ijassa-348	338	20	is	be	AUX
ijassa-348	338	21	huge	huge	ADJ
ijassa-348	338	22	the	the	DET
ijassa-348	338	23	time	time	NOUN
ijassa-348	338	24	span	span	NOUN
ijassa-348	338	25	is	be	AUX
ijassa-348	338	26	relatively	relatively	ADV
ijassa-348	338	27	huge	huge	ADJ
ijassa-348	338	28	in	in	ADP
ijassa-348	338	29	comparison	comparison	NOUN
ijassa-348	338	30	to	to	ADP
ijassa-348	338	31	the	the	DET
ijassa-348	338	32	nsl	nsl	NOUN
ijassa-348	338	33	-	-	PUNCT
ijassa-348	338	34	kdd	kdd	PROPN
ijassa-348	338	35	dataset	dataset	NOUN
ijassa-348	338	36	.	.	PUNCT
ijassa-348	339	1	figs	fig	NOUN
ijassa-348	339	2	6	6	NUM
ijassa-348	339	3	-	-	SYM
ijassa-348	339	4	9	9	NUM
ijassa-348	339	5	represent	represent	VERB
ijassa-348	339	6	the	the	DET
ijassa-348	339	7	graphical	graphical	ADJ
ijassa-348	339	8	accuracies	accuracy	NOUN
ijassa-348	339	9	using	use	VERB
ijassa-348	339	10	svm	svm	PROPN
ijassa-348	339	11	,	,	PUNCT
ijassa-348	339	12	ldc	ldc	PROPN
ijassa-348	339	13	,	,	PUNCT
ijassa-348	339	14	qdc	qdc	PROPN
ijassa-348	339	15	and	and	CCONJ
ijassa-348	339	16	wma	wma	NOUN
ijassa-348	339	17	managers	manager	NOUN
ijassa-348	339	18	for	for	ADP
ijassa-348	339	19	the	the	DET
ijassa-348	339	20	unsw	unsw	PROPN
ijassa-348	339	21	-	-	PUNCT
ijassa-348	339	22	nb	nb	NOUN
ijassa-348	339	23	dataset	dataset	NOUN
ijassa-348	339	24	respectively	respectively	ADV
ijassa-348	339	25	.	.	PUNCT
ijassa-348	340	1	it	it	PRON
ijassa-348	340	2	is	be	AUX
ijassa-348	340	3	inferred	infer	VERB
ijassa-348	340	4	that	that	SCONJ
ijassa-348	340	5	svm	svm	PROPN
ijassa-348	340	6	performs	perform	VERB
ijassa-348	340	7	poorly	poorly	ADV
ijassa-348	340	8	for	for	ADP
ijassa-348	340	9	attacks	attack	NOUN
ijassa-348	340	10	namely	namely	ADV
ijassa-348	340	11	reconnaissance	reconnaissance	NOUN
ijassa-348	340	12	,	,	PUNCT
ijassa-348	340	13	dos	do	NOUN
ijassa-348	340	14	and	and	CCONJ
ijassa-348	340	15	analysis	analysis	NOUN
ijassa-348	340	16	with	with	ADP
ijassa-348	340	17	accuracy	accuracy	NOUN
ijassa-348	340	18	of	of	ADP
ijassa-348	340	19	35	35	NUM
ijassa-348	340	20	%	%	NOUN
ijassa-348	340	21	,	,	PUNCT
ijassa-348	340	22	29	29	NUM
ijassa-348	340	23	%	%	NOUN
ijassa-348	340	24	and	and	CCONJ
ijassa-348	340	25	40	40	NUM
ijassa-348	340	26	%	%	NOUN
ijassa-348	340	27	respectively	respectively	ADV
ijassa-348	340	28	.	.	PUNCT
ijassa-348	341	1	ldc	ldc	PROPN
ijassa-348	341	2	performs	perform	VERB
ijassa-348	341	3	poorly	poorly	ADV
ijassa-348	341	4	for	for	ADP
ijassa-348	341	5	attacks	attack	NOUN
ijassa-348	341	6	such	such	ADJ
ijassa-348	341	7	as	as	ADP
ijassa-348	341	8	shellcode	shellcode	NOUN
ijassa-348	341	9	,	,	PUNCT
ijassa-348	341	10	reconnaissance	reconnaissance	NOUN
ijassa-348	341	11	,	,	PUNCT
ijassa-348	341	12	dos	do	NOUN
ijassa-348	341	13	and	and	CCONJ
ijassa-348	341	14	fuzzers	fuzzer	NOUN
ijassa-348	341	15	with	with	ADP
ijassa-348	341	16	accuracy	accuracy	NOUN
ijassa-348	341	17	of	of	ADP
ijassa-348	341	18	40%,36%,38	40%,36%,38	NUM
ijassa-348	341	19	%	%	NOUN
ijassa-348	341	20	and	and	CCONJ
ijassa-348	341	21	28	28	NUM
ijassa-348	341	22	%	%	NOUN
ijassa-348	341	23	respectively	respectively	ADV
ijassa-348	341	24	.	.	PUNCT
ijassa-348	342	1	qdc	qdc	PROPN
ijassa-348	342	2	performs	perform	VERB
ijassa-348	342	3	poorly	poorly	ADV
ijassa-348	342	4	for	for	ADP
ijassa-348	342	5	backdoor	backdoor	NOUN
ijassa-348	342	6	,	,	PUNCT
ijassa-348	342	7	analysis	analysis	NOUN
ijassa-348	342	8	,	,	PUNCT
ijassa-348	342	9	exploits	exploit	NOUN
ijassa-348	342	10	,	,	PUNCT
ijassa-348	342	11	shellcode	shellcode	NOUN
ijassa-348	342	12	and	and	CCONJ
ijassa-348	342	13	reconnaissance	reconnaissance	NOUN
ijassa-348	342	14	with	with	ADP
ijassa-348	342	15	accuracy	accuracy	NOUN
ijassa-348	342	16	of	of	ADP
ijassa-348	342	17	34%,33%,30	34%,33%,30	NUM
ijassa-348	342	18	%	%	NOUN
ijassa-348	342	19	and	and	CCONJ
ijassa-348	342	20	33	33	NUM
ijassa-348	342	21	%	%	NOUN
ijassa-348	342	22	respectively	respectively	ADV
ijassa-348	342	23	whereas	whereas	SCONJ
ijassa-348	342	24	wma	wma	NOUN
ijassa-348	342	25	results	result	NOUN
ijassa-348	342	26	in	in	ADP
ijassa-348	342	27	an	an	DET
ijassa-348	342	28	average	average	ADJ
ijassa-348	342	29	accuracy	accuracy	NOUN
ijassa-348	342	30	for	for	ADP
ijassa-348	342	31	reconnaissance	reconnaissance	NOUN
ijassa-348	342	32	,	,	PUNCT
ijassa-348	342	33	exploits	exploit	NOUN
ijassa-348	342	34	,	,	PUNCT
ijassa-348	342	35	shellcode	shellcode	NOUN
ijassa-348	342	36	and	and	CCONJ
ijassa-348	342	37	fuzzers	fuzzer	NOUN
ijassa-348	342	38	more	more	ADJ
ijassa-348	342	39	than	than	ADP
ijassa-348	342	40	50	50	NUM
ijassa-348	342	41	%	%	NOUN
ijassa-348	342	42	which	which	PRON
ijassa-348	342	43	is	be	AUX
ijassa-348	342	44	a	a	DET
ijassa-348	342	45	considerable	considerable	ADJ
ijassa-348	342	46	increase	increase	NOUN
ijassa-348	342	47	in	in	ADP
ijassa-348	342	48	comparison	comparison	NOUN
ijassa-348	342	49	to	to	ADP
ijassa-348	342	50	base	base	NOUN
ijassa-348	342	51	classifiers	classifier	NOUN
ijassa-348	342	52	and	and	CCONJ
ijassa-348	342	53	other	other	ADJ
ijassa-348	342	54	class	class	NOUN
ijassa-348	342	55	labels	label	NOUN
ijassa-348	342	56	also	also	ADV
ijassa-348	342	57	produce	produce	VERB
ijassa-348	342	58	higher	high	ADJ
ijassa-348	342	59	accuracy	accuracy	NOUN
ijassa-348	342	60	.	.	PUNCT
ijassa-348	343	1	table	table	NOUN
ijassa-348	343	2	7	7	NUM
ijassa-348	343	3	attributes	attribute	NOUN
ijassa-348	343	4	retrieved	retrieve	VERB
ijassa-348	343	5	after	after	ADP
ijassa-348	343	6	dimensionality	dimensionality	NOUN
ijassa-348	343	7	reduction	reduction	NOUN
ijassa-348	343	8	using	use	VERB
ijassa-348	343	9	pca	pca	NOUN
ijassa-348	343	10	in	in	ADP
ijassa-348	343	11	unswnb	unswnb	ADJ
ijassa-348	343	12	dataset	dataset	NOUN
ijassa-348	343	13	.	.	PUNCT
ijassa-348	344	1	21	21	NUM
ijassa-348	344	2	attributes	attribute	NOUN
ijassa-348	344	3	selected	select	VERB
ijassa-348	344	4	after	after	ADP
ijassa-348	344	5	pca	pca	PROPN
ijassa-348	344	6	i	i	PROPN
ijassa-348	344	7	d	d	PROPN
ijassa-348	344	8	,	,	PUNCT
ijassa-348	344	9	dur	dur	PROPN
ijassa-348	344	10	,	,	PUNCT
ijassa-348	344	11	service	service	NOUN
ijassa-348	344	12	,	,	PUNCT
ijassa-348	344	13	state	state	NOUN
ijassa-348	344	14	,	,	PUNCT
ijassa-348	344	15	spkts	spkts	NOUN
ijassa-348	344	16	,	,	PUNCT
ijassa-348	344	17	dpkts	dpkts	NOUN
ijassa-348	344	18	,	,	PUNCT
ijassa-348	344	19	sbytes	sbyte	NOUN
ijassa-348	344	20	,	,	PUNCT
ijassa-348	344	21	dbytes	dbyte	NOUN
ijassa-348	344	22	,	,	PUNCT
ijassa-348	344	23	rate	rate	NOUN
ijassa-348	344	24	,	,	PUNCT
ijassa-348	344	25	sttl	sttl	PROPN
ijassa-348	344	26	,	,	PUNCT
ijassa-348	344	27	dttl	dttl	PROPN
ijassa-348	344	28	,	,	PUNCT
ijassa-348	344	29	sload	sload	NOUN
ijassa-348	344	30	,	,	PUNCT
ijassa-348	344	31	dload	dload	NOUN
ijassa-348	344	32	,	,	PUNCT
ijassa-348	344	33	sloss	sloss	NOUN
ijassa-348	344	34	,	,	PUNCT
ijassa-348	344	35	dloss	dloss	NOUN
ijassa-348	344	36	,	,	PUNCT
ijassa-348	344	37	sinpkt	sinpkt	NOUN
ijassa-348	344	38	,	,	PUNCT
ijassa-348	344	39	dinpkt	dinpkt	ADJ
ijassa-348	344	40	,	,	PUNCT
ijassa-348	344	41	sjit	sjit	NOUN
ijassa-348	344	42	,	,	PUNCT
ijassa-348	344	43	djit	djit	PROPN
ijassa-348	344	44	,	,	PUNCT
ijassa-348	344	45	swin	swin	PROPN
ijassa-348	344	46	,	,	PUNCT
ijassa-348	344	47	stcpb	stcpb	NOUN
ijassa-348	344	48	,	,	PUNCT
ijassa-348	344	49	dtcpb	dtcpb	VERB
ijassa-348	344	50	,	,	PUNCT
ijassa-348	344	51	dwin	dwin	PROPN
ijassa-348	344	52	,	,	PUNCT
ijassa-348	344	53	tcprtt	tcprtt	ADJ
ijassa-348	344	54	,	,	PUNCT
ijassa-348	344	55	synack	synack	ADJ
ijassa-348	344	56	,	,	PUNCT
ijassa-348	344	57	ackdat	ackdat	NOUN
ijassa-348	344	58	,	,	PUNCT
ijassa-348	344	59	smean	smean	ADJ
ijassa-348	344	60	,	,	PUNCT
ijassa-348	344	61	dmean	dmean	ADJ
ijassa-348	344	62	,	,	PUNCT
ijassa-348	344	63	trans	tran	NOUN
ijassa-348	344	64	depth	depth	NOUN
ijassa-348	344	65	,	,	PUNCT
ijassa-348	344	66	response	response	NOUN
ijassa-348	344	67	body	body	NOUN
ijassa-348	344	68	len	len	VERB
ijassa-348	344	69	,	,	PUNCT
ijassa-348	344	70	ct	ct	PROPN
ijassa-348	344	71	srv	srv	PROPN
ijassa-348	344	72	src	src	PROPN
ijassa-348	344	73	,	,	PUNCT
ijassa-348	344	74	ct	ct	PROPN
ijassa-348	344	75	state	state	PROPN
ijassa-348	344	76	ttl	ttl	PROPN
ijassa-348	344	77	,	,	PUNCT
ijassa-348	344	78	ct	ct	PROPN
ijassa-348	344	79	dst	dst	PROPN
ijassa-348	344	80	ltm	ltm	PROPN
ijassa-348	344	81	,	,	PUNCT
ijassa-348	344	82	ct	ct	NUM
ijassa-348	344	83	src	src	NOUN
ijassa-348	344	84	dport	dport	NOUN
ijassa-348	344	85	ltm	ltm	PROPN
ijassa-348	344	86	,	,	PUNCT
ijassa-348	344	87	ct	ct	PROPN
ijassa-348	344	88	dst	dst	PROPN
ijassa-348	344	89	sport	sport	PROPN
ijassa-348	344	90	ltm	ltm	PROPN
ijassa-348	344	91	,	,	PUNCT
ijassa-348	344	92	is-2	is-2	NOUN
ijassa-348	344	93	login	login	NOUN
ijassa-348	344	94	,	,	PUNCT
ijassa-348	344	95	ct	ct	PROPN
ijassa-348	344	96	2	2	NUM
ijassa-348	344	97	cmd	cmd	NOUN
ijassa-348	344	98	,	,	PUNCT
ijassa-348	344	99	ctflw	ctflw	PROPN
ijassa-348	344	100	4	4	NUM
ijassa-348	344	101	mthd	mthd	ADJ
ijassa-348	344	102	,	,	PUNCT
ijassa-348	344	103	ct	ct	NUM
ijassa-348	344	104	src	src	NOUN
ijassa-348	344	105	ltm	ltm	PROPN
ijassa-348	344	106	,	,	PUNCT
ijassa-348	344	107	ct	ct	PROPN
ijassa-348	344	108	srv	srv	PROPN
ijassa-348	344	109	dst	dst	PROPN
ijassa-348	344	110	,	,	PUNCT
ijassa-348	344	111	is	be	AUX
ijassa-348	344	112	sm	sm	PROPN
ijassa-348	344	113	ips	ips	NOUN
ijassa-348	344	114	ports	port	NOUN
ijassa-348	344	115	.	.	PUNCT
ijassa-348	345	1	advances	advance	NOUN
ijassa-348	345	2	in	in	ADP
ijassa-348	345	3	systems	system	NOUN
ijassa-348	345	4	science	science	NOUN
ijassa-348	345	5	and	and	CCONJ
ijassa-348	345	6	application	application	NOUN
ijassa-348	345	7	(	(	PUNCT
ijassa-348	345	8	2016	2016	NUM
ijassa-348	345	9	)	)	PUNCT
ijassa-348	346	1	vol.16	vol.16	PROPN
ijassa-348	346	2	no.2	no.2	PROPN
ijassa-348	346	3	33	33	NUM
ijassa-348	346	4	table	table	NOUN
ijassa-348	346	5	8	8	NUM
ijassa-348	346	6	experimental	experimental	ADJ
ijassa-348	346	7	results	result	NOUN
ijassa-348	346	8	using	use	VERB
ijassa-348	346	9	svm	svm	ADJ
ijassa-348	346	10	manager	manager	NOUN
ijassa-348	346	11	in	in	ADP
ijassa-348	346	12	unsw	unsw	PROPN
ijassa-348	346	13	-	-	PUNCT
ijassa-348	346	14	nb	nb	NOUN
ijassa-348	346	15	dataset	dataset	NOUN
ijassa-348	346	16	manager	manager	NOUN
ijassa-348	346	17	normal	normal	ADJ
ijassa-348	346	18	analysis	analysis	NOUN
ijassa-348	346	19	backdoor	backdoor	NOUN
ijassa-348	346	20	reconna	reconna	ADJ
ijassa-348	346	21	issance	issance	NOUN
ijassa-348	346	22	exploits	exploit	VERB
ijassa-348	346	23	fuzzers	fuzzer	NOUN
ijassa-348	346	24	generic	generic	ADJ
ijassa-348	346	25	dos	do	NOUN
ijassa-348	346	26	shellcode	shellcode	PROPN
ijassa-348	346	27	svm	svm	PROPN
ijassa-348	346	28	1	1	NUM
ijassa-348	346	29	99.94	99.94	NUM
ijassa-348	346	30	66.21	66.21	NUM
ijassa-348	346	31	59.14	59.14	NUM
ijassa-348	346	32	90.28	90.28	NUM
ijassa-348	346	33	86.76	86.76	NUM
ijassa-348	346	34	91.35	91.35	NUM
ijassa-348	346	35	99.61	99.61	NUM
ijassa-348	346	36	91.35	91.35	NUM
ijassa-348	346	37	91.26	91.26	NUM
ijassa-348	346	38	svm	svm	NOUN
ijassa-348	346	39	2	2	NUM
ijassa-348	346	40	51.66	51.66	NUM
ijassa-348	346	41	60.94	60.94	NUM
ijassa-348	346	42	50.81	50.81	NUM
ijassa-348	346	43	71.89	71.89	NUM
ijassa-348	346	44	64.86	64.86	NUM
ijassa-348	346	45	98.33	98.33	NUM
ijassa-348	346	46	84.06	84.06	NUM
ijassa-348	346	47	76.66	76.66	NUM
ijassa-348	346	48	86.66	86.66	NUM
ijassa-348	346	49	svm	svm	NOUN
ijassa-348	346	50	3	3	NUM
ijassa-348	346	51	69.14	69.14	NUM
ijassa-348	346	52	93.79	93.79	NUM
ijassa-348	346	53	45.81	45.81	NUM
ijassa-348	346	54	48.21	48.21	NUM
ijassa-348	346	55	56.17	56.17	NUM
ijassa-348	346	56	99.72	99.72	NUM
ijassa-348	346	57	86.89	86.89	NUM
ijassa-348	346	58	56	56	NUM
ijassa-348	346	59	78.87	78.87	NUM
ijassa-348	346	60	svm	svm	VERB
ijassa-348	346	61	4	4	NUM
ijassa-348	346	62	99.03	99.03	NUM
ijassa-348	346	63	39.28	39.28	NUM
ijassa-348	346	64	45.95	45.95	NUM
ijassa-348	346	65	60.79	60.79	NUM
ijassa-348	346	66	87.76	87.76	NUM
ijassa-348	346	67	53.05	53.05	NUM
ijassa-348	346	68	98.4	98.4	NUM
ijassa-348	346	69	70.2	70.2	NUM
ijassa-348	346	70	27.83	27.83	NUM
ijassa-348	346	71	svm	svm	NOUN
ijassa-348	346	72	5	5	NUM
ijassa-348	346	73	93.76	93.76	NUM
ijassa-348	346	74	43.33	43.33	NUM
ijassa-348	346	75	65.51	65.51	NUM
ijassa-348	346	76	25.97	25.97	NUM
ijassa-348	346	77	47.81	47.81	NUM
ijassa-348	346	78	57.88	57.88	NUM
ijassa-348	346	79	72.64	72.64	NUM
ijassa-348	346	80	58.62	58.62	NUM
ijassa-348	346	81	43.13	43.13	NUM
ijassa-348	346	82	svm	svm	PROPN
ijassa-348	346	83	6	6	NUM
ijassa-348	346	84	82.34	82.34	NUM
ijassa-348	346	85	34.76	34.76	NUM
ijassa-348	346	86	62	62	NUM
ijassa-348	346	87	30.39	30.39	NUM
ijassa-348	346	88	52.22	52.22	NUM
ijassa-348	346	89	63.57	63.57	NUM
ijassa-348	346	90	87.37	87.37	NUM
ijassa-348	346	91	80.76	80.76	NUM
ijassa-348	346	92	73.22	73.22	NUM
ijassa-348	346	93	svm	svm	NOUN
ijassa-348	346	94	7	7	NUM
ijassa-348	346	95	99.55	99.55	NUM
ijassa-348	346	96	31.33	31.33	NUM
ijassa-348	346	97	47.36	47.36	NUM
ijassa-348	346	98	33.55	33.55	NUM
ijassa-348	346	99	75.49	75.49	NUM
ijassa-348	346	100	42.67	42.67	NUM
ijassa-348	346	101	99.26	99.26	NUM
ijassa-348	346	102	47.42	47.42	NUM
ijassa-348	346	103	42	42	NUM
ijassa-348	346	104	svm	svm	ADJ
ijassa-348	346	105	8	8	NUM
ijassa-348	346	106	96.19	96.19	NUM
ijassa-348	346	107	32.15	32.15	NUM
ijassa-348	346	108	47.07	47.07	NUM
ijassa-348	346	109	33.4	33.4	NUM
ijassa-348	346	110	52.54	52.54	NUM
ijassa-348	346	111	33.84	33.84	NUM
ijassa-348	346	112	98.83	98.83	NUM
ijassa-348	346	113	28.95	28.95	NUM
ijassa-348	346	114	34.18	34.18	NUM
ijassa-348	346	115	svm	svm	NOUN
ijassa-348	346	116	9	9	NUM
ijassa-348	346	117	99.65	99.65	NUM
ijassa-348	346	118	79.63	79.63	NUM
ijassa-348	346	119	63.73	63.73	NUM
ijassa-348	346	120	70.78	70.78	NUM
ijassa-348	346	121	87.38	87.38	NUM
ijassa-348	346	122	96.28	96.28	NUM
ijassa-348	346	123	99.65	99.65	NUM
ijassa-348	346	124	93.97	93.97	NUM
ijassa-348	346	125	87.51	87.51	NUM
ijassa-348	346	126	table	table	NOUN
ijassa-348	346	127	9	9	NUM
ijassa-348	346	128	experimental	experimental	ADJ
ijassa-348	346	129	results	result	NOUN
ijassa-348	346	130	using	use	VERB
ijassa-348	346	131	ldc	ldc	PROPN
ijassa-348	346	132	manager	manager	NOUN
ijassa-348	346	133	in	in	ADP
ijassa-348	346	134	unsw	unsw	PROPN
ijassa-348	346	135	-	-	PUNCT
ijassa-348	346	136	nb	nb	NOUN
ijassa-348	346	137	dataset	dataset	NOUN
ijassa-348	346	138	manager	manager	NOUN
ijassa-348	346	139	normal	normal	ADJ
ijassa-348	346	140	analysis	analysis	NOUN
ijassa-348	346	141	backdoor	backdoor	NOUN
ijassa-348	346	142	reconnaissance	reconnaissance	NOUN
ijassa-348	346	143	exploits	exploit	VERB
ijassa-348	346	144	fuzzers	fuzzer	NOUN
ijassa-348	346	145	generic	generic	ADJ
ijassa-348	346	146	dos	do	NOUN
ijassa-348	346	147	shellcode	shellcode	PROPN
ijassa-348	346	148	ldc	ldc	PROPN
ijassa-348	346	149	1	1	NUM
ijassa-348	346	150	87.91	87.91	NUM
ijassa-348	346	151	53.49	53.49	NUM
ijassa-348	346	152	54.15	54.15	NUM
ijassa-348	346	153	75.07	75.07	NUM
ijassa-348	346	154	67.88	67.88	NUM
ijassa-348	346	155	63.91	63.91	NUM
ijassa-348	346	156	99.09	99.09	NUM
ijassa-348	346	157	71.95	71.95	NUM
ijassa-348	346	158	39.61	39.61	NUM
ijassa-348	346	159	ldc	ldc	PROPN
ijassa-348	346	160	2	2	NUM
ijassa-348	346	161	86.41	86.41	NUM
ijassa-348	346	162	50.22	50.22	NUM
ijassa-348	346	163	52.95	52.95	NUM
ijassa-348	346	164	36.07	36.07	NUM
ijassa-348	346	165	84.34	84.34	NUM
ijassa-348	346	166	55.66	55.66	NUM
ijassa-348	346	167	94.71	94.71	NUM
ijassa-348	346	168	54.16	54.16	NUM
ijassa-348	346	169	84.94	84.94	NUM
ijassa-348	346	170	ldc	ldc	PROPN
ijassa-348	346	171	3	3	NUM
ijassa-348	346	172	98.67	98.67	NUM
ijassa-348	346	173	75.14	75.14	NUM
ijassa-348	346	174	62.59	62.59	NUM
ijassa-348	346	175	36.84	36.84	NUM
ijassa-348	346	176	65.45	65.45	NUM
ijassa-348	346	177	56.08	56.08	NUM
ijassa-348	346	178	92.36	92.36	NUM
ijassa-348	346	179	38.67	38.67	NUM
ijassa-348	346	180	50	50	NUM
ijassa-348	346	181	ldc	ldc	PROPN
ijassa-348	346	182	4	4	NUM
ijassa-348	346	183	97.59	97.59	NUM
ijassa-348	346	184	80.11	80.11	NUM
ijassa-348	346	185	92.3	92.3	NUM
ijassa-348	346	186	32.98	32.98	NUM
ijassa-348	346	187	65.79	65.79	NUM
ijassa-348	346	188	42.87	42.87	NUM
ijassa-348	346	189	99.26	99.26	NUM
ijassa-348	346	190	46.25	46.25	NUM
ijassa-348	346	191	41.37	41.37	NUM
ijassa-348	346	192	ldc	ldc	PROPN
ijassa-348	346	193	5	5	NUM
ijassa-348	346	194	96.59	96.59	NUM
ijassa-348	346	195	89.2	89.2	NUM
ijassa-348	346	196	90.44	90.44	NUM
ijassa-348	346	197	50.89	50.89	NUM
ijassa-348	346	198	84.26	84.26	NUM
ijassa-348	346	199	47.55	47.55	NUM
ijassa-348	346	200	63.43	63.43	NUM
ijassa-348	346	201	35.26	35.26	NUM
ijassa-348	346	202	36.74	36.74	NUM
ijassa-348	346	203	ldc	ldc	PROPN
ijassa-348	346	204	6	6	NUM
ijassa-348	346	205	94.22	94.22	NUM
ijassa-348	346	206	76.56	76.56	NUM
ijassa-348	346	207	85.66	85.66	NUM
ijassa-348	346	208	36	36	NUM
ijassa-348	346	209	75.44	75.44	NUM
ijassa-348	346	210	94.59	94.59	NUM
ijassa-348	346	211	80.82	80.82	NUM
ijassa-348	346	212	48.21	48.21	NUM
ijassa-348	346	213	66.99	66.99	NUM
ijassa-348	346	214	ldc	ldc	PROPN
ijassa-348	346	215	7	7	NUM
ijassa-348	346	216	87.56	87.56	NUM
ijassa-348	346	217	44.7	44.7	NUM
ijassa-348	346	218	59.53	59.53	NUM
ijassa-348	346	219	37.39	37.39	NUM
ijassa-348	346	220	55.4	55.4	NUM
ijassa-348	346	221	28.41	28.41	NUM
ijassa-348	346	222	99.57	99.57	NUM
ijassa-348	346	223	38.39	38.39	NUM
ijassa-348	346	224	43.27	43.27	NUM
ijassa-348	346	225	table	table	NOUN
ijassa-348	346	226	10	10	NUM
ijassa-348	346	227	experimental	experimental	ADJ
ijassa-348	346	228	results	result	NOUN
ijassa-348	346	229	using	use	VERB
ijassa-348	346	230	qdc	qdc	PROPN
ijassa-348	346	231	manager	manager	NOUN
ijassa-348	346	232	in	in	ADP
ijassa-348	346	233	unsw	unsw	PROPN
ijassa-348	346	234	-	-	PUNCT
ijassa-348	346	235	nb	nb	NOUN
ijassa-348	346	236	dataset	dataset	NOUN
ijassa-348	346	237	manager	manager	NOUN
ijassa-348	346	238	normal	normal	ADJ
ijassa-348	346	239	analysis	analysis	NOUN
ijassa-348	346	240	backdoor	backdoor	NOUN
ijassa-348	346	241	reconnaissance	reconnaissance	NOUN
ijassa-348	346	242	exploits	exploit	VERB
ijassa-348	346	243	fuzzers	fuzzer	NOUN
ijassa-348	346	244	generic	generic	ADJ
ijassa-348	346	245	dos	do	NOUN
ijassa-348	346	246	shellcode	shellcode	PROPN
ijassa-348	346	247	qdc	qdc	VERB
ijassa-348	346	248	1	1	NUM
ijassa-348	346	249	82.11	82.11	NUM
ijassa-348	346	250	46.41	46.41	NUM
ijassa-348	347	1	77.41	77.41	NUM
ijassa-348	348	1	26.38	26.38	NUM
ijassa-348	348	2	48.33	48.33	NUM
ijassa-348	348	3	50	50	NUM
ijassa-348	348	4	65.21	65.21	NUM
ijassa-348	348	5	67.1	67.1	NUM
ijassa-348	348	6	67.16	67.16	NUM
ijassa-348	348	7	qdc	qdc	PROPN
ijassa-348	348	8	2	2	NUM
ijassa-348	348	9	97.86	97.86	NUM
ijassa-348	348	10	28.67	28.67	NUM
ijassa-348	348	11	25.03	25.03	NUM
ijassa-348	348	12	35.8	35.8	NUM
ijassa-348	348	13	60.34	60.34	NUM
ijassa-348	348	14	57.06	57.06	NUM
ijassa-348	348	15	80.68	80.68	NUM
ijassa-348	348	16	75.49	75.49	NUM
ijassa-348	348	17	58.76	58.76	NUM
ijassa-348	348	18	qdc	qdc	PROPN
ijassa-348	348	19	3	3	NUM
ijassa-348	348	20	86.85	86.85	NUM
ijassa-348	348	21	32.98	32.98	NUM
ijassa-348	348	22	34.03	34.03	NUM
ijassa-348	348	23	33.42	33.42	NUM
ijassa-348	348	24	83.7	83.7	NUM
ijassa-348	348	25	60.34	60.34	NUM
ijassa-348	348	26	59.06	59.06	NUM
ijassa-348	348	27	85.68	85.68	NUM
ijassa-348	348	28	50	50	NUM
ijassa-348	348	29	qdc	qdc	NOUN
ijassa-348	348	30	4	4	NUM
ijassa-348	348	31	91.73	91.73	NUM
ijassa-348	348	32	43.43	43.43	NUM
ijassa-348	348	33	52.76	52.76	NUM
ijassa-348	348	34	36.34	36.34	NUM
ijassa-348	348	35	46.27	46.27	NUM
ijassa-348	348	36	51.37	51.37	NUM
ijassa-348	348	37	94.03	94.03	NUM
ijassa-348	348	38	77.36	77.36	NUM
ijassa-348	348	39	30.17	30.17	NUM
ijassa-348	348	40	qdc	qdc	PROPN
ijassa-348	348	41	5	5	NUM
ijassa-348	348	42	84.21	84.21	NUM
ijassa-348	348	43	83.33	83.33	NUM
ijassa-348	348	44	63.37	63.37	NUM
ijassa-348	348	45	42.19	42.19	NUM
ijassa-348	348	46	51.76	51.76	NUM
ijassa-348	348	47	45.83	45.83	NUM
ijassa-348	348	48	80.94	80.94	NUM
ijassa-348	348	49	74.15	74.15	NUM
ijassa-348	348	50	42.23	42.23	NUM
ijassa-348	348	51	qdc	qdc	PROPN
ijassa-348	348	52	6	6	NUM
ijassa-348	348	53	81.24	81.24	NUM
ijassa-348	348	54	35.82	35.82	NUM
ijassa-348	348	55	81.51	81.51	NUM
ijassa-348	348	56	24.4	24.4	NUM
ijassa-348	348	57	52.8	52.8	NUM
ijassa-348	348	58	53.95	53.95	NUM
ijassa-348	348	59	87.59	87.59	NUM
ijassa-348	348	60	81.92	81.92	NUM
ijassa-348	348	61	41.95	41.95	NUM
ijassa-348	348	62	qdc	qdc	PROPN
ijassa-348	348	63	7	7	NUM
ijassa-348	348	64	88.39	88.39	NUM
ijassa-348	348	65	40.52	40.52	NUM
ijassa-348	348	66	78.94	78.94	NUM
ijassa-348	348	67	49.49	49.49	NUM
ijassa-348	348	68	38.58	38.58	NUM
ijassa-348	348	69	54.98	54.98	NUM
ijassa-348	348	70	87.58	87.58	NUM
ijassa-348	348	71	68.85	68.85	NUM
ijassa-348	348	72	64.19	64.19	NUM
ijassa-348	348	73	qdc	qdc	PROPN
ijassa-348	348	74	8	8	NUM
ijassa-348	348	75	87.48	87.48	NUM
ijassa-348	348	76	44.45	44.45	NUM
ijassa-348	348	77	58.95	58.95	NUM
ijassa-348	348	78	35.43	35.43	NUM
ijassa-348	348	79	54.54	54.54	NUM
ijassa-348	348	80	66.31	66.31	NUM
ijassa-348	348	81	81.54	81.54	NUM
ijassa-348	348	82	91.66	91.66	NUM
ijassa-348	348	83	50.63	50.63	NUM
ijassa-348	348	84	qdc	qdc	NOUN
ijassa-348	348	85	9	9	NUM
ijassa-348	348	86	88.25	88.25	NUM
ijassa-348	348	87	46.7	46.7	NUM
ijassa-348	348	88	63.85	63.85	NUM
ijassa-348	348	89	36.72	36.72	NUM
ijassa-348	348	90	62.33	62.33	NUM
ijassa-348	348	91	62.83	62.83	NUM
ijassa-348	348	92	96.43	96.43	NUM
ijassa-348	348	93	82.48	82.48	NUM
ijassa-348	348	94	53.56	53.56	NUM
ijassa-348	348	95	table	table	NOUN
ijassa-348	348	96	11	11	NUM
ijassa-348	348	97	experimental	experimental	ADJ
ijassa-348	348	98	results	result	NOUN
ijassa-348	348	99	using	use	VERB
ijassa-348	348	100	wma	wma	NOUN
ijassa-348	348	101	manager	manager	NOUN
ijassa-348	348	102	in	in	ADP
ijassa-348	348	103	unsw	unsw	PROPN
ijassa-348	348	104	-	-	PUNCT
ijassa-348	348	105	nb	nb	NOUN
ijassa-348	348	106	dataset	dataset	NOUN
ijassa-348	348	107	manager	manager	NOUN
ijassa-348	348	108	normal	normal	ADJ
ijassa-348	348	109	analysis	analysis	NOUN
ijassa-348	348	110	backdoor	backdoor	NOUN
ijassa-348	348	111	reconnaissance	reconnaissance	NOUN
ijassa-348	348	112	exploits	exploit	VERB
ijassa-348	348	113	fuzzers	fuzzer	NOUN
ijassa-348	348	114	generic	generic	ADJ
ijassa-348	348	115	dos	do	NOUN
ijassa-348	348	116	shellcode	shellcode	PROPN
ijassa-348	348	117	wma1	wma1	NOUN
ijassa-348	348	118	99.99	99.99	NUM
ijassa-348	348	119	78.41	78.41	NUM
ijassa-348	348	120	86.71	86.71	NUM
ijassa-348	348	121	65.97	65.97	NUM
ijassa-348	348	122	76.66	76.66	NUM
ijassa-348	348	123	91.35	91.35	NUM
ijassa-348	348	124	91.85	91.85	NUM
ijassa-348	348	125	58.94	58.94	NUM
ijassa-348	348	126	86.66	86.66	NUM
ijassa-348	348	127	wma1	wma1	NOUN
ijassa-348	348	128	98.51	98.51	NUM
ijassa-348	348	129	51.95	51.95	NUM
ijassa-348	348	130	64.71	64.71	NUM
ijassa-348	348	131	71.89	71.89	NUM
ijassa-348	348	132	68.84	68.84	NUM
ijassa-348	348	133	98.33	98.33	NUM
ijassa-348	348	134	90.88	90.88	NUM
ijassa-348	348	135	50.46	50.46	NUM
ijassa-348	348	136	91.26	91.26	NUM
ijassa-348	348	137	wma	wma	NOUN
ijassa-348	348	138	3	3	NUM
ijassa-348	348	139	88.93	88.93	NUM
ijassa-348	348	140	52	52	NUM
ijassa-348	348	141	62.09	62.09	NUM
ijassa-348	348	142	48.21	48.21	NUM
ijassa-348	348	143	71.14	71.14	NUM
ijassa-348	348	144	99.72	99.72	NUM
ijassa-348	348	145	81.45	81.45	NUM
ijassa-348	348	146	90.75	90.75	NUM
ijassa-348	348	147	78.87	78.87	NUM
ijassa-348	348	148	wma	wma	NOUN
ijassa-348	348	149	4	4	NUM
ijassa-348	348	150	99.41	99.41	NUM
ijassa-348	348	151	50.8	50.8	NUM
ijassa-348	348	152	100	100	NUM
ijassa-348	348	153	60.79	60.79	NUM
ijassa-348	348	154	73.97	73.97	NUM
ijassa-348	348	155	53.05	53.05	NUM
ijassa-348	348	156	97.65	97.65	NUM
ijassa-348	348	157	‘	'	PUNCT
ijassa-348	348	158	100	100	NUM
ijassa-348	348	159	57.83	57.83	NUM
ijassa-348	348	160	wma	wma	NOUN
ijassa-348	348	161	5	5	NUM
ijassa-348	348	162	93.2	93.2	NUM
ijassa-348	348	163	67.32	67.32	NUM
ijassa-348	348	164	84.61	84.61	NUM
ijassa-348	348	165	90.28	90.28	NUM
ijassa-348	348	166	59.9	59.9	NUM
ijassa-348	348	167	57.88	57.88	NUM
ijassa-348	348	168	90.19	90.19	NUM
ijassa-348	348	169	67.67	67.67	NUM
ijassa-348	348	170	53.13	53.13	NUM
ijassa-348	348	171	wma	wma	NOUN
ijassa-348	348	172	6	6	NUM
ijassa-348	348	173	54	54	NUM
ijassa-348	348	174	100	100	NUM
ijassa-348	348	175	50	50	NUM
ijassa-348	348	176	60.39	60.39	NUM
ijassa-348	348	177	46.84	46.84	NUM
ijassa-348	348	178	63.57	63.57	NUM
ijassa-348	348	179	84.32	84.32	NUM
ijassa-348	348	180	68.8	68.8	NUM
ijassa-348	348	181	73.22	73.22	NUM
ijassa-348	348	182	wma	wma	NOUN
ijassa-348	348	183	7	7	NUM
ijassa-348	348	184	95.25	95.25	NUM
ijassa-348	348	185	55.89	55.89	NUM
ijassa-348	348	186	57.09	57.09	NUM
ijassa-348	348	187	63.55	63.55	NUM
ijassa-348	348	188	62.67	62.67	NUM
ijassa-348	348	189	42.67	42.67	NUM
ijassa-348	348	190	99.15	99.15	NUM
ijassa-348	348	191	58.8	58.8	NUM
ijassa-348	348	192	62	62	NUM
ijassa-348	348	193	wma	wma	NOUN
ijassa-348	348	194	8	8	NUM
ijassa-348	348	195	81.71	81.71	NUM
ijassa-348	348	196	74.72	74.72	NUM
ijassa-348	348	197	50.22	50.22	NUM
ijassa-348	348	198	63.4	63.4	NUM
ijassa-348	348	199	88.21	88.21	NUM
ijassa-348	348	200	63.84	63.84	NUM
ijassa-348	348	201	91.4	91.4	NUM
ijassa-348	348	202	69.34	69.34	NUM
ijassa-348	348	203	64.18	64.18	NUM
ijassa-348	348	204	wma	wma	NOUN
ijassa-348	348	205	9	9	NUM
ijassa-348	348	206	99.21	99.21	NUM
ijassa-348	348	207	58.67	58.67	NUM
ijassa-348	348	208	65.62	65.62	NUM
ijassa-348	348	209	70.78	70.78	NUM
ijassa-348	348	210	87.92	87.92	NUM
ijassa-348	348	211	96.28	96.28	NUM
ijassa-348	348	212	84.13	84.13	NUM
ijassa-348	348	213	64.96	64.96	NUM
ijassa-348	348	214	87.51	87.51	NUM
ijassa-348	348	215	table	table	NOUN
ijassa-348	348	216	12	12	NUM
ijassa-348	348	217	elapsed	elapse	VERB
ijassa-348	348	218	time	time	NOUN
ijassa-348	348	219	for	for	SCONJ
ijassa-348	348	220	managers	manager	NOUN
ijassa-348	348	221	in	in	ADP
ijassa-348	348	222	unsw	unsw	PROPN
ijassa-348	348	223	-	-	PUNCT
ijassa-348	348	224	nb	nb	PROPN
ijassa-348	348	225	data	data	PROPN
ijassa-348	348	226	set	set	NOUN
ijassa-348	348	227	managers	manager	NOUN
ijassa-348	348	228	elapsed	elapse	VERB
ijassa-348	348	229	time	time	NOUN
ijassa-348	348	230	(	(	PUNCT
ijassa-348	348	231	secs	secs	X
ijassa-348	348	232	)	)	PUNCT
ijassa-348	348	233	svm	svm	VERB
ijassa-348	348	234	464.82	464.82	NUM
ijassa-348	348	235	s	s	PART
ijassa-348	348	236	ldc	ldc	PROPN
ijassa-348	348	237	485.56	485.56	NUM
ijassa-348	348	238	s	s	NOUN
ijassa-348	348	239	qdc	qdc	PROPN
ijassa-348	348	240	622.50	622.50	NUM
ijassa-348	348	241	s	s	PART
ijassa-348	348	242	wma	wma	NOUN
ijassa-348	348	243	751.61	751.61	NUM
ijassa-348	348	244	s	s	NOUN
ijassa-348	348	245	34	34	NUM
ijassa-348	348	246	sumaiya	sumaiya	NOUN
ijassa-348	348	247	thaseen	thaseen	PROPN
ijassa-348	348	248	and	and	CCONJ
ijassa-348	348	249	ch.aswani	ch.aswani	X
ijassa-348	349	1	kumar	kumar	PROPN
ijassa-348	349	2	:	:	PUNCT
ijassa-348	349	3	intrusion	intrusion	NOUN
ijassa-348	349	4	detection	detection	NOUN
ijassa-348	349	5	model	model	NOUN
ijassa-348	349	6	using	use	VERB
ijassa-348	349	7	pca	pca	PROPN
ijassa-348	349	8	and	and	CCONJ
ijassa-348	349	9	...	...	PUNCT
ijassa-348	349	10	fig	fig	NOUN
ijassa-348	349	11	.	.	PUNCT
ijassa-348	350	1	6	6	NUM
ijassa-348	350	2	proposed	propose	VERB
ijassa-348	350	3	intrusion	intrusion	NOUN
ijassa-348	350	4	detection	detection	NOUN
ijassa-348	350	5	model	model	NOUN
ijassa-348	350	6	fig	fig	NOUN
ijassa-348	350	7	.	.	PUNCT
ijassa-348	351	1	7	7	NUM
ijassa-348	351	2	proposed	propose	VERB
ijassa-348	351	3	intrusion	intrusion	NOUN
ijassa-348	351	4	detection	detection	NOUN
ijassa-348	351	5	model	model	NOUN
ijassa-348	351	6	fig	fig	NOUN
ijassa-348	351	7	.	.	PUNCT
ijassa-348	351	8	8	8	NUM
ijassa-348	351	9	proposed	propose	VERB
ijassa-348	351	10	intrusion	intrusion	NOUN
ijassa-348	351	11	detection	detection	NOUN
ijassa-348	351	12	model	model	NOUN
ijassa-348	351	13	advances	advance	NOUN
ijassa-348	351	14	in	in	ADP
ijassa-348	351	15	systems	system	NOUN
ijassa-348	351	16	science	science	NOUN
ijassa-348	351	17	and	and	CCONJ
ijassa-348	351	18	application	application	NOUN
ijassa-348	351	19	(	(	PUNCT
ijassa-348	351	20	2016	2016	NUM
ijassa-348	351	21	)	)	PUNCT
ijassa-348	352	1	vol.16	vol.16	PROPN
ijassa-348	352	2	no.2	no.2	PROPN
ijassa-348	352	3	35	35	NUM
ijassa-348	352	4	fig	fig	NOUN
ijassa-348	352	5	.	.	PUNCT
ijassa-348	353	1	9	9	NUM
ijassa-348	353	2	proposed	propose	VERB
ijassa-348	353	3	intrusion	intrusion	NOUN
ijassa-348	353	4	detection	detection	NOUN
ijassa-348	353	5	model	model	VERB
ijassa-348	353	6	5.3	5.3	NUM
ijassa-348	353	7	discussion	discussion	NOUN
ijassa-348	353	8	in	in	ADP
ijassa-348	353	9	this	this	DET
ijassa-348	353	10	paper	paper	NOUN
ijassa-348	353	11	,	,	PUNCT
ijassa-348	353	12	we	we	PRON
ijassa-348	353	13	have	have	AUX
ijassa-348	353	14	analyzed	analyze	VERB
ijassa-348	353	15	the	the	DET
ijassa-348	353	16	accuracy	accuracy	NOUN
ijassa-348	353	17	of	of	ADP
ijassa-348	353	18	the	the	DET
ijassa-348	353	19	base	base	NOUN
ijassa-348	353	20	and	and	CCONJ
ijassa-348	353	21	ensemble	ensemble	ADJ
ijassa-348	353	22	classifier	classifier	NOUN
ijassa-348	353	23	managers	manager	NOUN
ijassa-348	353	24	.	.	PUNCT
ijassa-348	354	1	it	it	PRON
ijassa-348	354	2	is	be	AUX
ijassa-348	354	3	to	to	PART
ijassa-348	354	4	be	be	AUX
ijassa-348	354	5	noted	note	VERB
ijassa-348	354	6	that	that	SCONJ
ijassa-348	354	7	the	the	DET
ijassa-348	354	8	ensemble	ensemble	ADJ
ijassa-348	354	9	managers	manager	NOUN
ijassa-348	354	10	accuracy	accuracy	NOUN
ijassa-348	354	11	is	be	AUX
ijassa-348	354	12	relatively	relatively	ADV
ijassa-348	354	13	high	high	ADJ
ijassa-348	354	14	in	in	ADP
ijassa-348	354	15	comparison	comparison	NOUN
ijassa-348	354	16	to	to	ADP
ijassa-348	354	17	the	the	DET
ijassa-348	354	18	base	base	NOUN
ijassa-348	354	19	classifiers	classifier	NOUN
ijassa-348	354	20	and	and	CCONJ
ijassa-348	354	21	managers	manager	NOUN
ijassa-348	354	22	accuracies	accuracy	NOUN
ijassa-348	354	23	are	be	AUX
ijassa-348	354	24	above	above	ADP
ijassa-348	354	25	99	99	NUM
ijassa-348	354	26	%	%	NOUN
ijassa-348	354	27	for	for	ADP
ijassa-348	354	28	nsl	nsl	NOUN
ijassa-348	354	29	-	-	PUNCT
ijassa-348	354	30	kdd	kdd	NOUN
ijassa-348	354	31	dataset	dataset	NOUN
ijassa-348	354	32	and	and	CCONJ
ijassa-348	354	33	the	the	DET
ijassa-348	354	34	range	range	NOUN
ijassa-348	354	35	is	be	AUX
ijassa-348	354	36	from	from	ADP
ijassa-348	354	37	88	88	NUM
ijassa-348	354	38	-	-	SYM
ijassa-348	354	39	100	100	NUM
ijassa-348	354	40	%	%	NOUN
ijassa-348	354	41	for	for	ADP
ijassa-348	354	42	unsw	unsw	NOUN
ijassa-348	354	43	-	-	PUNCT
ijassa-348	354	44	nb	nb	NOUN
ijassa-348	354	45	dataset	dataset	NOUN
ijassa-348	354	46	.	.	PUNCT
ijassa-348	355	1	the	the	DET
ijassa-348	355	2	variation	variation	NOUN
ijassa-348	355	3	is	be	AUX
ijassa-348	355	4	due	due	ADJ
ijassa-348	355	5	to	to	ADP
ijassa-348	355	6	the	the	DET
ijassa-348	355	7	imbalance	imbalance	NOUN
ijassa-348	355	8	of	of	ADP
ijassa-348	355	9	samples	sample	NOUN
ijassa-348	355	10	in	in	ADP
ijassa-348	355	11	the	the	DET
ijassa-348	355	12	dataset	dataset	NOUN
ijassa-348	355	13	for	for	ADP
ijassa-348	355	14	the	the	DET
ijassa-348	355	15	classes	class	NOUN
ijassa-348	355	16	exploits	exploit	NOUN
ijassa-348	355	17	,	,	PUNCT
ijassa-348	355	18	shellcode	shellcode	NOUN
ijassa-348	355	19	and	and	CCONJ
ijassa-348	355	20	reconnaissance	reconnaissance	NOUN
ijassa-348	355	21	.	.	PUNCT
ijassa-348	356	1	the	the	DET
ijassa-348	356	2	reason	reason	NOUN
ijassa-348	356	3	for	for	ADP
ijassa-348	356	4	the	the	DET
ijassa-348	356	5	high	high	ADJ
ijassa-348	356	6	accuracy	accuracy	NOUN
ijassa-348	356	7	in	in	ADP
ijassa-348	356	8	both	both	CCONJ
ijassa-348	356	9	the	the	DET
ijassa-348	356	10	datasets	dataset	NOUN
ijassa-348	356	11	are	be	AUX
ijassa-348	356	12	1	1	NUM
ijassa-348	356	13	)	)	PUNCT
ijassa-348	356	14	proper	proper	ADJ
ijassa-348	356	15	selection	selection	NOUN
ijassa-348	356	16	of	of	ADP
ijassa-348	356	17	training	training	NOUN
ijassa-348	356	18	data	datum	NOUN
ijassa-348	356	19	2	2	NUM
ijassa-348	356	20	)	)	PUNCT
ijassa-348	356	21	well	well	ADV
ijassa-348	356	22	-	-	PUNCT
ijassa-348	356	23	selected	select	VERB
ijassa-348	356	24	parameters	parameter	NOUN
ijassa-348	356	25	in	in	ADP
ijassa-348	356	26	wma	wma	NOUN
ijassa-348	356	27	technique	technique	NOUN
ijassa-348	356	28	namely	namely	ADV
ijassa-348	356	29	the	the	DET
ijassa-348	356	30	learning	learn	VERB
ijassa-348	356	31	factor	factor	NOUN
ijassa-348	356	32	and	and	CCONJ
ijassa-348	356	33	weight	weight	NOUN
ijassa-348	356	34	assignment	assignment	NOUN
ijassa-348	356	35	wi	wi	PROPN
ijassa-348	356	36	for	for	ADP
ijassa-348	356	37	each	each	DET
ijassa-348	356	38	sample	sample	NOUN
ijassa-348	356	39	.	.	PUNCT
ijassa-348	357	1	3	3	X
ijassa-348	357	2	)	)	PUNCT
ijassa-348	357	3	misclassification	misclassification	NOUN
ijassa-348	357	4	error	error	NOUN
ijassa-348	357	5	is	be	AUX
ijassa-348	357	6	reduced	reduce	VERB
ijassa-348	357	7	due	due	ADP
ijassa-348	357	8	to	to	ADP
ijassa-348	357	9	weight	weight	NOUN
ijassa-348	357	10	assignment	assignment	NOUN
ijassa-348	357	11	to	to	ADP
ijassa-348	357	12	each	each	DET
ijassa-348	357	13	sample	sample	NOUN
ijassa-348	357	14	.	.	PUNCT
ijassa-348	358	1	there	there	PRON
ijassa-348	358	2	were	be	VERB
ijassa-348	358	3	poor	poor	ADJ
ijassa-348	358	4	results	result	NOUN
ijassa-348	358	5	obtained	obtain	VERB
ijassa-348	358	6	for	for	ADP
ijassa-348	358	7	ldc	ldc	PROPN
ijassa-348	358	8	and	and	CCONJ
ijassa-348	358	9	qdc	qdc	NOUN
ijassa-348	358	10	manager	manager	NOUN
ijassa-348	358	11	only	only	ADV
ijassa-348	358	12	in	in	ADP
ijassa-348	358	13	the	the	DET
ijassa-348	358	14	unsw	unsw	PROPN
ijassa-348	358	15	-	-	PUNCT
ijassa-348	358	16	nb	nb	NOUN
ijassa-348	358	17	dataset	dataset	NOUN
ijassa-348	358	18	.	.	PUNCT
ijassa-348	359	1	but	but	CCONJ
ijassa-348	359	2	svm	svm	PROPN
ijassa-348	359	3	performs	perform	VERB
ijassa-348	359	4	well	well	ADV
ijassa-348	359	5	for	for	ADP
ijassa-348	359	6	nsl	nsl	NOUN
ijassa-348	359	7	-	-	PUNCT
ijassa-348	359	8	kdd	kdd	NOUN
ijassa-348	359	9	dataset	dataset	NOUN
ijassa-348	359	10	and	and	CCONJ
ijassa-348	359	11	in	in	ADP
ijassa-348	359	12	specific	specific	ADJ
ijassa-348	359	13	for	for	ADP
ijassa-348	359	14	class	class	NOUN
ijassa-348	359	15	labels	label	NOUN
ijassa-348	359	16	normal	normal	ADJ
ijassa-348	359	17	,	,	PUNCT
ijassa-348	359	18	fuzzers	fuzzer	NOUN
ijassa-348	359	19	,	,	PUNCT
ijassa-348	359	20	generic	generic	ADJ
ijassa-348	359	21	and	and	CCONJ
ijassa-348	359	22	dos	do	NOUN
ijassa-348	359	23	in	in	ADP
ijassa-348	359	24	the	the	DET
ijassa-348	359	25	unsw	unsw	PROPN
ijassa-348	359	26	-	-	PUNCT
ijassa-348	359	27	nb	nb	NOUN
ijassa-348	359	28	dataset	dataset	NOUN
ijassa-348	359	29	.	.	PUNCT
ijassa-348	360	1	svm	svm	PROPN
ijassa-348	360	2	produced	produce	VERB
ijassa-348	360	3	poor	poor	ADJ
ijassa-348	360	4	accuracy	accuracy	NOUN
ijassa-348	360	5	of	of	ADP
ijassa-348	360	6	85	85	NUM
ijassa-348	360	7	%	%	NOUN
ijassa-348	360	8	for	for	ADP
ijassa-348	360	9	dos	do	NOUN
ijassa-348	360	10	and	and	CCONJ
ijassa-348	360	11	qdc	qdc	VERB
ijassa-348	360	12	produced	produce	VERB
ijassa-348	360	13	poor	poor	ADJ
ijassa-348	360	14	accuracy	accuracy	NOUN
ijassa-348	360	15	of	of	ADP
ijassa-348	360	16	91	91	NUM
ijassa-348	360	17	%	%	NOUN
ijassa-348	360	18	and	and	CCONJ
ijassa-348	360	19	95	95	NUM
ijassa-348	360	20	%	%	NOUN
ijassa-348	360	21	for	for	ADP
ijassa-348	360	22	u2r	u2r	NOUN
ijassa-348	360	23	and	and	CCONJ
ijassa-348	360	24	r2l	r2l	NOUN
ijassa-348	360	25	attacks	attack	NOUN
ijassa-348	360	26	in	in	ADP
ijassa-348	360	27	the	the	DET
ijassa-348	360	28	nsl	nsl	NOUN
ijassa-348	360	29	-	-	PUNCT
ijassa-348	360	30	kdd	kdd	PROPN
ijassa-348	360	31	dataset	dataset	NOUN
ijassa-348	360	32	.	.	PUNCT
ijassa-348	361	1	the	the	DET
ijassa-348	361	2	accuracy	accuracy	NOUN
ijassa-348	361	3	was	be	AUX
ijassa-348	361	4	as	as	ADV
ijassa-348	361	5	low	low	ADJ
ijassa-348	361	6	as	as	ADP
ijassa-348	361	7	49.49	49.49	NUM
ijassa-348	361	8	%	%	NOUN
ijassa-348	361	9	for	for	ADP
ijassa-348	361	10	reconnaissance	reconnaissance	NOUN
ijassa-348	361	11	class	class	NOUN
ijassa-348	361	12	label	label	NOUN
ijassa-348	361	13	because	because	SCONJ
ijassa-348	361	14	the	the	DET
ijassa-348	361	15	test	test	NOUN
ijassa-348	361	16	data	data	NOUN
ijassa-348	361	17	contains	contain	VERB
ijassa-348	361	18	samples	sample	NOUN
ijassa-348	361	19	that	that	PRON
ijassa-348	361	20	were	be	AUX
ijassa-348	361	21	not	not	PART
ijassa-348	361	22	utilized	utilize	VERB
ijassa-348	361	23	while	while	SCONJ
ijassa-348	361	24	training	train	VERB
ijassa-348	361	25	the	the	DET
ijassa-348	361	26	classifiers	classifier	NOUN
ijassa-348	361	27	.	.	PUNCT
ijassa-348	362	1	thus	thus	ADV
ijassa-348	362	2	in	in	ADP
ijassa-348	362	3	comparison	comparison	NOUN
ijassa-348	362	4	to	to	ADP
ijassa-348	362	5	wma	wma	PROPN
ijassa-348	362	6	,	,	PUNCT
ijassa-348	362	7	the	the	DET
ijassa-348	362	8	base	base	NOUN
ijassa-348	362	9	classifiers	classifier	NOUN
ijassa-348	362	10	svm	svm	VERB
ijassa-348	362	11	,	,	PUNCT
ijassa-348	362	12	ldc	ldc	PROPN
ijassa-348	362	13	and	and	CCONJ
ijassa-348	362	14	qdc	qdc	PROPN
ijassa-348	362	15	perform	perform	VERB
ijassa-348	362	16	poorly	poorly	ADV
ijassa-348	362	17	and	and	CCONJ
ijassa-348	362	18	also	also	ADV
ijassa-348	362	19	with	with	ADP
ijassa-348	362	20	increase	increase	NOUN
ijassa-348	362	21	in	in	ADP
ijassa-348	362	22	elapsed	elapse	VERB
ijassa-348	362	23	time	time	NOUN
ijassa-348	362	24	.	.	PUNCT
ijassa-348	363	1	the	the	DET
ijassa-348	363	2	constraint	constraint	NOUN
ijassa-348	363	3	on	on	ADP
ijassa-348	363	4	the	the	DET
ijassa-348	363	5	weights	weight	NOUN
ijassa-348	363	6	generated	generate	VERB
ijassa-348	363	7	by	by	ADP
ijassa-348	363	8	wma	wma	NOUN
ijassa-348	363	9	also	also	ADV
ijassa-348	363	10	is	be	AUX
ijassa-348	363	11	a	a	DET
ijassa-348	363	12	major	major	ADJ
ijassa-348	363	13	factor	factor	NOUN
ijassa-348	363	14	in	in	ADP
ijassa-348	363	15	improving	improve	VERB
ijassa-348	363	16	the	the	DET
ijassa-348	363	17	accuracy	accuracy	NOUN
ijassa-348	363	18	of	of	ADP
ijassa-348	363	19	the	the	DET
ijassa-348	363	20	intrusion	intrusion	NOUN
ijassa-348	363	21	detection	detection	NOUN
ijassa-348	363	22	model	model	NOUN
ijassa-348	363	23	.	.	PUNCT
ijassa-348	364	1	the	the	DET
ijassa-348	364	2	weights	weight	NOUN
ijassa-348	364	3	can	can	AUX
ijassa-348	364	4	lie	lie	VERB
ijassa-348	364	5	between	between	ADP
ijassa-348	364	6	0	0	NUM
ijassa-348	364	7	and	and	CCONJ
ijassa-348	364	8	1	1	NUM
ijassa-348	364	9	,	,	PUNCT
ijassa-348	364	10	the	the	PRON
ijassa-348	364	11	higher	high	ADJ
ijassa-348	364	12	the	the	DET
ijassa-348	364	13	value	value	NOUN
ijassa-348	364	14	the	the	PRON
ijassa-348	364	15	more	more	ADV
ijassa-348	364	16	is	be	AUX
ijassa-348	364	17	the	the	DET
ijassa-348	364	18	possibility	possibility	NOUN
ijassa-348	364	19	of	of	ADP
ijassa-348	364	20	correctly	correctly	ADV
ijassa-348	364	21	predicted	predict	VERB
ijassa-348	364	22	.	.	PUNCT
ijassa-348	365	1	thus	thus	ADV
ijassa-348	365	2	this	this	DET
ijassa-348	365	3	model	model	NOUN
ijassa-348	365	4	can	can	AUX
ijassa-348	365	5	be	be	AUX
ijassa-348	365	6	extended	extend	VERB
ijassa-348	365	7	to	to	PART
ijassa-348	365	8	include	include	VERB
ijassa-348	365	9	other	other	ADJ
ijassa-348	365	10	ensemble	ensemble	ADJ
ijassa-348	365	11	techniques	technique	NOUN
ijassa-348	365	12	but	but	CCONJ
ijassa-348	365	13	it	it	PRON
ijassa-348	365	14	should	should	AUX
ijassa-348	365	15	consider	consider	VERB
ijassa-348	365	16	the	the	DET
ijassa-348	365	17	time	time	NOUN
ijassa-348	365	18	complexity	complexity	NOUN
ijassa-348	365	19	as	as	SCONJ
ijassa-348	365	20	many	many	ADJ
ijassa-348	365	21	ensemble	ensemble	ADJ
ijassa-348	365	22	approaches	approach	NOUN
ijassa-348	365	23	complete	complete	VERB
ijassa-348	365	24	the	the	DET
ijassa-348	365	25	task	task	NOUN
ijassa-348	365	26	in	in	ADP
ijassa-348	365	27	extremely	extremely	ADV
ijassa-348	365	28	long	long	ADJ
ijassa-348	365	29	time	time	NOUN
ijassa-348	365	30	.	.	PUNCT
ijassa-348	366	1	36	36	NUM
ijassa-348	366	2	sumaiya	sumaiya	NOUN
ijassa-348	366	3	thaseen	thaseen	PROPN
ijassa-348	366	4	and	and	CCONJ
ijassa-348	366	5	ch.aswani	ch.aswani	X
ijassa-348	367	1	kumar	kumar	PROPN
ijassa-348	367	2	:	:	PUNCT
ijassa-348	367	3	intrusion	intrusion	NOUN
ijassa-348	367	4	detection	detection	NOUN
ijassa-348	367	5	model	model	NOUN
ijassa-348	367	6	using	use	VERB
ijassa-348	367	7	pca	pca	PROPN
ijassa-348	367	8	and	and	CCONJ
ijassa-348	367	9	...	...	PUNCT
ijassa-348	367	10	6	6	NUM
ijassa-348	367	11	conclusions	conclusion	NOUN
ijassa-348	367	12	the	the	DET
ijassa-348	367	13	classification	classification	NOUN
ijassa-348	367	14	accuracy	accuracy	NOUN
ijassa-348	367	15	can	can	AUX
ijassa-348	367	16	be	be	AUX
ijassa-348	367	17	improved	improve	VERB
ijassa-348	367	18	with	with	ADP
ijassa-348	367	19	minimal	minimal	ADJ
ijassa-348	367	20	elapsed	elapse	VERB
ijassa-348	367	21	time	time	NOUN
ijassa-348	367	22	by	by	ADP
ijassa-348	367	23	integrating	integrate	VERB
ijassa-348	367	24	opinions	opinion	NOUN
ijassa-348	367	25	from	from	ADP
ijassa-348	367	26	multiple	multiple	ADJ
ijassa-348	367	27	managers	manager	NOUN
ijassa-348	367	28	into	into	ADP
ijassa-348	367	29	single	single	ADJ
ijassa-348	367	30	using	use	VERB
ijassa-348	367	31	an	an	DET
ijassa-348	367	32	ensemble	ensemble	ADJ
ijassa-348	367	33	approach	approach	NOUN
ijassa-348	367	34	.	.	PUNCT
ijassa-348	368	1	we	we	PRON
ijassa-348	368	2	have	have	AUX
ijassa-348	368	3	deployed	deploy	VERB
ijassa-348	368	4	weighted	weight	VERB
ijassa-348	368	5	majority	majority	NOUN
ijassa-348	368	6	voting	voting	NOUN
ijassa-348	368	7	to	to	PART
ijassa-348	368	8	integrate	integrate	VERB
ijassa-348	368	9	results	result	NOUN
ijassa-348	368	10	from	from	ADP
ijassa-348	368	11	different	different	ADJ
ijassa-348	368	12	managers	manager	NOUN
ijassa-348	368	13	.	.	PUNCT
ijassa-348	369	1	the	the	DET
ijassa-348	369	2	three	three	NUM
ijassa-348	369	3	base	base	NOUN
ijassa-348	369	4	classifiers	classifier	NOUN
ijassa-348	369	5	namely	namely	ADV
ijassa-348	369	6	svm	svm	PROPN
ijassa-348	369	7	,	,	PUNCT
ijassa-348	369	8	ldc	ldc	PROPN
ijassa-348	369	9	and	and	CCONJ
ijassa-348	369	10	qdc	qdc	PROPN
ijassa-348	369	11	were	be	AUX
ijassa-348	369	12	experimentally	experimentally	ADV
ijassa-348	369	13	compared	compare	VERB
ijassa-348	369	14	using	use	VERB
ijassa-348	369	15	two	two	NUM
ijassa-348	369	16	different	different	ADJ
ijassa-348	369	17	datasets	dataset	NOUN
ijassa-348	369	18	namely	namely	ADV
ijassa-348	369	19	nsl	nsl	NOUN
ijassa-348	369	20	-	-	PUNCT
ijassa-348	369	21	kdd	kdd	PROPN
ijassa-348	369	22	and	and	CCONJ
ijassa-348	369	23	unswnb	unswnb	ADJ
ijassa-348	369	24	.	.	PUNCT
ijassa-348	370	1	thus	thus	ADV
ijassa-348	370	2	wma	wma	NOUN
ijassa-348	370	3	results	result	NOUN
ijassa-348	370	4	in	in	ADP
ijassa-348	370	5	good	good	ADJ
ijassa-348	370	6	accuracy	accuracy	NOUN
ijassa-348	370	7	for	for	ADP
ijassa-348	370	8	both	both	CCONJ
ijassa-348	370	9	the	the	DET
ijassa-348	370	10	datasets	dataset	NOUN
ijassa-348	370	11	.	.	PUNCT
ijassa-348	371	1	the	the	DET
ijassa-348	371	2	accuracy	accuracy	NOUN
ijassa-348	371	3	improvement	improvement	NOUN
ijassa-348	371	4	is	be	AUX
ijassa-348	371	5	of	of	ADP
ijassa-348	371	6	1	1	NUM
ijassa-348	371	7	%	%	NOUN
ijassa-348	371	8	in	in	ADP
ijassa-348	371	9	comparison	comparison	NOUN
ijassa-348	371	10	to	to	ADP
ijassa-348	371	11	base	base	NOUN
ijassa-348	371	12	classifiers	classifier	NOUN
ijassa-348	371	13	.	.	PUNCT
ijassa-348	372	1	thus	thus	ADV
ijassa-348	372	2	the	the	DET
ijassa-348	372	3	success	success	NOUN
ijassa-348	372	4	of	of	ADP
ijassa-348	372	5	the	the	DET
ijassa-348	372	6	model	model	NOUN
ijassa-348	372	7	is	be	AUX
ijassa-348	372	8	due	due	ADJ
ijassa-348	372	9	to	to	ADP
ijassa-348	372	10	the	the	DET
ijassa-348	372	11	generated	generate	VERB
ijassa-348	372	12	weights	weight	NOUN
ijassa-348	372	13	in	in	ADP
ijassa-348	372	14	wma	wma	NOUN
ijassa-348	372	15	which	which	PRON
ijassa-348	372	16	were	be	AUX
ijassa-348	372	17	tuned	tune	VERB
ijassa-348	372	18	by	by	ADP
ijassa-348	372	19	the	the	DET
ijassa-348	372	20	learning	learning	NOUN
ijassa-348	372	21	factor	factor	NOUN
ijassa-348	372	22	.	.	PUNCT
ijassa-348	373	1	thus	thus	ADV
ijassa-348	373	2	the	the	DET
ijassa-348	373	3	integration	integration	NOUN
ijassa-348	373	4	of	of	ADP
ijassa-348	373	5	base	base	NOUN
ijassa-348	373	6	classifiers	classifier	NOUN
ijassa-348	373	7	in	in	ADP
ijassa-348	373	8	to	to	ADP
ijassa-348	373	9	an	an	DET
ijassa-348	373	10	ensemble	ensemble	ADJ
ijassa-348	373	11	manager	manager	NOUN
ijassa-348	373	12	with	with	ADP
ijassa-348	373	13	wma	wma	NOUN
ijassa-348	373	14	will	will	AUX
ijassa-348	373	15	be	be	AUX
ijassa-348	373	16	very	very	ADV
ijassa-348	373	17	well	well	ADV
ijassa-348	373	18	utilized	utilize	VERB
ijassa-348	373	19	in	in	ADP
ijassa-348	373	20	intrusion	intrusion	NOUN
ijassa-348	373	21	detection	detection	NOUN
ijassa-348	373	22	as	as	SCONJ
ijassa-348	373	23	it	it	PRON
ijassa-348	373	24	has	have	AUX
ijassa-348	373	25	been	be	AUX
ijassa-348	373	26	tested	test	VERB
ijassa-348	373	27	on	on	ADP
ijassa-348	373	28	recent	recent	ADJ
ijassa-348	373	29	datasets	dataset	NOUN
ijassa-348	373	30	with	with	ADP
ijassa-348	373	31	modern	modern	ADJ
ijassa-348	373	32	day	day	NOUN
ijassa-348	373	33	attacks	attack	NOUN
ijassa-348	373	34	.	.	PUNCT
ijassa-348	374	1	the	the	DET
ijassa-348	374	2	other	other	ADJ
ijassa-348	374	3	improvement	improvement	NOUN
ijassa-348	374	4	in	in	ADP
ijassa-348	374	5	our	our	PRON
ijassa-348	374	6	approach	approach	NOUN
ijassa-348	374	7	is	be	AUX
ijassa-348	374	8	instead	instead	ADV
ijassa-348	374	9	of	of	ADP
ijassa-348	374	10	relying	rely	VERB
ijassa-348	374	11	on	on	ADP
ijassa-348	374	12	binary	binary	ADJ
ijassa-348	374	13	classification	classification	NOUN
ijassa-348	374	14	techniques	technique	NOUN
ijassa-348	374	15	,	,	PUNCT
ijassa-348	374	16	we	we	PRON
ijassa-348	374	17	have	have	AUX
ijassa-348	374	18	utilized	utilize	VERB
ijassa-348	374	19	multi	multi	ADJ
ijassa-348	374	20	class	class	NOUN
ijassa-348	374	21	classification	classification	NOUN
ijassa-348	374	22	for	for	ADP
ijassa-348	374	23	the	the	DET
ijassa-348	374	24	base	base	NOUN
ijassa-348	374	25	managers	manager	NOUN
ijassa-348	374	26	.	.	PUNCT
ijassa-348	375	1	in	in	ADP
ijassa-348	375	2	future	future	NOUN
ijassa-348	375	3	we	we	PRON
ijassa-348	375	4	can	can	AUX
ijassa-348	375	5	integrate	integrate	VERB
ijassa-348	375	6	optimization	optimization	NOUN
ijassa-348	375	7	techniques	technique	NOUN
ijassa-348	375	8	with	with	ADP
ijassa-348	375	9	wma	wma	NOUN
ijassa-348	375	10	to	to	PART
ijassa-348	375	11	tune	tune	VERB
ijassa-348	375	12	the	the	DET
ijassa-348	375	13	parameters	parameter	NOUN
ijassa-348	375	14	for	for	ADP
ijassa-348	375	15	generating	generate	VERB
ijassa-348	375	16	weights	weight	NOUN
ijassa-348	375	17	.	.	PUNCT
ijassa-348	376	1	references	reference	NOUN
ijassa-348	376	2	[	[	X
ijassa-348	376	3	1	1	NUM
ijassa-348	376	4	]	]	X
ijassa-348	376	5	khan	khan	PROPN
ijassa-348	376	6	l.	l.	PROPN
ijassa-348	376	7	,	,	PUNCT
ijassa-348	376	8	awad	awad	PROPN
ijassa-348	376	9	m.	m.	PROPN
ijassa-348	376	10	and	and	CCONJ
ijassa-348	376	11	thruaisingham	thruaisingham	PROPN
ijassa-348	376	12	b.	b.	PROPN
ijassa-348	376	13	(	(	PUNCT
ijassa-348	376	14	2007	2007	NUM
ijassa-348	376	15	)	)	PUNCT
ijassa-348	376	16	,	,	PUNCT
ijassa-348	376	17	“	"	PUNCT
ijassa-348	376	18	a	a	DET
ijassa-348	376	19	new	new	ADJ
ijassa-348	376	20	intrusion	intrusion	NOUN
ijassa-348	376	21	detection	detection	NOUN
ijassa-348	376	22	system	system	NOUN
ijassa-348	376	23	using	use	VERB
ijassa-348	376	24	support	support	NOUN
ijassa-348	376	25	vector	vector	NOUN
ijassa-348	376	26	machines	machine	NOUN
ijassa-348	376	27	and	and	CCONJ
ijassa-348	376	28	hierarchical	hierarchical	ADJ
ijassa-348	376	29	clustering	clustering	NOUN
ijassa-348	376	30	”	"	PUNCT
ijassa-348	376	31	,	,	PUNCT
ijassa-348	376	32	the	the	DET
ijassa-348	376	33	vldb	vldb	ADJ
ijassa-348	376	34	journal	journal	NOUN
ijassa-348	376	35	,	,	PUNCT
ijassa-348	376	36	vol	vol	NOUN
ijassa-348	376	37	.	.	PROPN
ijassa-348	376	38	16	16	NUM
ijassa-348	376	39	,	,	PUNCT
ijassa-348	376	40	pp	pp	ADJ
ijassa-348	376	41	.	.	PUNCT
ijassa-348	377	1	507	507	NUM
ijassa-348	377	2	-	-	SYM
ijassa-348	377	3	521	521	NUM
ijassa-348	377	4	.	.	PUNCT
ijassa-348	378	1	[	[	X
ijassa-348	378	2	2	2	X
ijassa-348	378	3	]	]	PUNCT
ijassa-348	378	4	wang	wang	PROPN
ijassa-348	378	5	g.	g.	PROPN
ijassa-348	378	6	,	,	PUNCT
ijassa-348	378	7	hao	hao	PROPN
ijassa-348	378	8	j.	j.	PROPN
ijassa-348	378	9	,	,	PUNCT
ijassa-348	378	10	ma	ma	PROPN
ijassa-348	378	11	j.	j.	PROPN
ijassa-348	378	12	and	and	CCONJ
ijassa-348	378	13	huang	huang	PROPN
ijassa-348	378	14	l.	l.	PROPN
ijassa-348	378	15	(	(	PUNCT
ijassa-348	378	16	2010	2010	NUM
ijassa-348	378	17	)	)	PUNCT
ijassa-348	378	18	,	,	PUNCT
ijassa-348	378	19	“	"	PUNCT
ijassa-348	378	20	a	a	DET
ijassa-348	378	21	new	new	ADJ
ijassa-348	378	22	approach	approach	NOUN
ijassa-348	378	23	to	to	ADP
ijassa-348	378	24	intrusion	intrusion	NOUN
ijassa-348	378	25	detection	detection	NOUN
ijassa-348	378	26	using	use	VERB
ijassa-348	378	27	artificial	artificial	ADJ
ijassa-348	378	28	neural	neural	ADJ
ijassa-348	378	29	networks	network	NOUN
ijassa-348	378	30	and	and	CCONJ
ijassa-348	378	31	fuzzy	fuzzy	ADJ
ijassa-348	378	32	clustering	clustering	NOUN
ijassa-348	378	33	,	,	PUNCT
ijassa-348	378	34	expert	expert	NOUN
ijassa-348	378	35	systems	system	NOUN
ijassa-348	378	36	with	with	ADP
ijassa-348	378	37	applications	application	NOUN
ijassa-348	378	38	”	"	PUNCT
ijassa-348	378	39	,	,	PUNCT
ijassa-348	378	40	vol	vol	NOUN
ijassa-348	378	41	.	.	PROPN
ijassa-348	378	42	37	37	NUM
ijassa-348	378	43	,	,	PUNCT
ijassa-348	378	44	pp	pp	ADJ
ijassa-348	378	45	.	.	PUNCT
ijassa-348	379	1	6225	6225	NUM
ijassa-348	379	2	-	-	SYM
ijassa-348	379	3	6232	6232	NUM
ijassa-348	379	4	.	.	PUNCT
ijassa-348	380	1	[	[	X
ijassa-348	380	2	3	3	X
ijassa-348	380	3	]	]	X
ijassa-348	380	4	sandhya	sandhya	PROPN
ijassa-348	380	5	peddabachigiri	peddabachigiri	PROPN
ijassa-348	380	6	,	,	PUNCT
ijassa-348	380	7	ajith	ajith	PROPN
ijassa-348	380	8	abraham	abraham	PROPN
ijassa-348	380	9	,	,	PUNCT
ijassa-348	380	10	crina	crina	PROPN
ijassa-348	380	11	grosan	grosan	NOUN
ijassa-348	380	12	and	and	CCONJ
ijassa-348	380	13	johnson	johnson	PROPN
ijassa-348	380	14	thomas	thomas	PROPN
ijassa-348	380	15	.	.	PUNCT
ijassa-348	381	1	(	(	PUNCT
ijassa-348	381	2	2007	2007	NUM
ijassa-348	381	3	)	)	PUNCT
ijassa-348	381	4	,	,	PUNCT
ijassa-348	381	5	“	"	PUNCT
ijassa-348	381	6	modeling	model	VERB
ijassa-348	381	7	intrusion	intrusion	NOUN
ijassa-348	381	8	detection	detection	NOUN
ijassa-348	381	9	system	system	NOUN
ijassa-348	381	10	using	use	VERB
ijassa-348	381	11	hybrid	hybrid	ADJ
ijassa-348	381	12	intelligent	intelligent	ADJ
ijassa-348	381	13	systems	system	NOUN
ijassa-348	381	14	”	"	PUNCT
ijassa-348	381	15	,	,	PUNCT
ijassa-348	381	16	journal	journal	NOUN
ijassa-348	381	17	of	of	ADP
ijassa-348	381	18	network	network	NOUN
ijassa-348	381	19	and	and	CCONJ
ijassa-348	381	20	computer	computer	NOUN
ijassa-348	381	21	applications	application	NOUN
ijassa-348	381	22	,	,	PUNCT
ijassa-348	381	23	vol	vol	NOUN
ijassa-348	381	24	.	.	PROPN
ijassa-348	381	25	30	30	NUM
ijassa-348	381	26	,	,	PUNCT
ijassa-348	381	27	pp	pp	ADJ
ijassa-348	381	28	.	.	PUNCT
ijassa-348	382	1	114	114	NUM
ijassa-348	382	2	-	-	SYM
ijassa-348	382	3	132	132	NUM
ijassa-348	382	4	.	.	PUNCT
ijassa-348	383	1	[	[	X
ijassa-348	383	2	4	4	X
ijassa-348	383	3	]	]	X
ijassa-348	383	4	lee	lee	PROPN
ijassa-348	383	5	w	w	PROPN
ijassa-348	383	6	,	,	PUNCT
ijassa-348	383	7	stolfo	stolfo	PROPN
ijassa-348	383	8	s	s	PART
ijassa-348	383	9	and	and	CCONJ
ijassa-348	383	10	mok	mok	PROPN
ijassa-348	383	11	k.	k.	PROPN
ijassa-348	383	12	(	(	PUNCT
ijassa-348	383	13	1999	1999	NUM
ijassa-348	383	14	)	)	PUNCT
ijassa-348	383	15	,	,	PUNCT
ijassa-348	383	16	“	"	PUNCT
ijassa-348	383	17	a	a	DET
ijassa-348	383	18	data	data	NOUN
ijassa-348	383	19	mining	mining	NOUN
ijassa-348	383	20	framework	framework	NOUN
ijassa-348	383	21	for	for	ADP
ijassa-348	383	22	building	building	NOUN
ijassa-348	383	23	intrusion	intrusion	NOUN
ijassa-348	383	24	detection	detection	NOUN
ijassa-348	383	25	model	model	NOUN
ijassa-348	383	26	”	"	PUNCT
ijassa-348	383	27	,	,	PUNCT
ijassa-348	383	28	in	in	ADP
ijassa-348	383	29	:	:	PUNCT
ijassa-348	383	30	proc	proc	NOUN
ijassa-348	383	31	.	.	PROPN
ijassa-348	384	1	of	of	ADP
ijassa-348	384	2	ieee	ieee	NOUN
ijassa-348	384	3	symposium	symposium	NOUN
ijassa-348	384	4	on	on	ADP
ijassa-348	384	5	security	security	NOUN
ijassa-348	384	6	and	and	CCONJ
ijassa-348	384	7	privacy	privacy	NOUN
ijassa-348	384	8	,	,	PUNCT
ijassa-348	384	9	pp	pp	X
ijassa-348	384	10	.	.	PUNCT
ijassa-348	385	1	120	120	NUM
ijassa-348	385	2	-	-	SYM
ijassa-348	385	3	32	32	NUM
ijassa-348	385	4	.	.	PUNCT
ijassa-348	386	1	[	[	X
ijassa-348	386	2	5	5	NUM
ijassa-348	386	3	]	]	PUNCT
ijassa-348	386	4	a.	a.	NOUN
ijassa-348	386	5	kausar	kausar	PROPN
ijassa-348	386	6	,	,	PUNCT
ijassa-348	386	7	m.	m.	NOUN
ijassa-348	386	8	ishtiaq	ishtiaq	PROPN
ijassa-348	386	9	,	,	PUNCT
ijassa-348	386	10	m.a.jaffar	m.a.jaffar	ADJ
ijassa-348	386	11	and	and	CCONJ
ijassa-348	386	12	a.m.mirza	a.m.mirza	ADJ
ijassa-348	386	13	.	.	PUNCT
ijassa-348	387	1	(	(	PUNCT
ijassa-348	387	2	2010	2010	NUM
ijassa-348	387	3	)	)	PUNCT
ijassa-348	387	4	,	,	PUNCT
ijassa-348	387	5	“	"	PUNCT
ijassa-348	387	6	optimization	optimization	NOUN
ijassa-348	387	7	of	of	ADP
ijassa-348	387	8	ensemble	ensemble	ADJ
ijassa-348	387	9	based	base	VERB
ijassa-348	387	10	decision	decision	NOUN
ijassa-348	387	11	using	use	VERB
ijassa-348	387	12	pso	pso	NOUN
ijassa-348	387	13	”	"	PUNCT
ijassa-348	387	14	,	,	PUNCT
ijassa-348	387	15	in	in	ADP
ijassa-348	387	16	:	:	PUNCT
ijassa-348	387	17	proceedings	proceeding	NOUN
ijassa-348	387	18	of	of	ADP
ijassa-348	387	19	the	the	DET
ijassa-348	387	20	world	world	NOUN
ijassa-348	387	21	congress	congress	PROPN
ijassa-348	387	22	on	on	ADP
ijassa-348	387	23	engineering	engineering	NOUN
ijassa-348	387	24	,	,	PUNCT
ijassa-348	387	25	vol	vol	NOUN
ijassa-348	387	26	.	.	PROPN
ijassa-348	387	27	10	10	NUM
ijassa-348	387	28	.	.	PUNCT
ijassa-348	388	1	[	[	X
ijassa-348	388	2	6	6	NUM
ijassa-348	388	3	]	]	X
ijassa-348	388	4	tsang	tsang	PROPN
ijassa-348	388	5	c.h	c.h	PROPN
ijassa-348	388	6	.	.	PROPN
ijassa-348	388	7	,	,	PUNCT
ijassa-348	388	8	kwong	kwong	PROPN
ijassa-348	388	9	s.	s.	PROPN
ijassa-348	388	10	and	and	CCONJ
ijassa-348	388	11	wang	wang	PROPN
ijassa-348	388	12	h.	h.	PROPN
ijassa-348	388	13	(	(	PUNCT
ijassa-348	388	14	2007	2007	NUM
ijassa-348	388	15	)	)	PUNCT
ijassa-348	388	16	,	,	PUNCT
ijassa-348	388	17	“	"	PUNCT
ijassa-348	388	18	genetic	genetic	ADJ
ijassa-348	388	19	fuzzy	fuzzy	ADJ
ijassa-348	388	20	rule	rule	NOUN
ijassa-348	388	21	mining	mining	NOUN
ijassa-348	388	22	approach	approach	NOUN
ijassa-348	388	23	and	and	CCONJ
ijassa-348	388	24	evaluation	evaluation	NOUN
ijassa-348	388	25	of	of	ADP
ijassa-348	388	26	feature	feature	NOUN
ijassa-348	388	27	selection	selection	NOUN
ijassa-348	388	28	techniques	technique	NOUN
ijassa-348	388	29	for	for	ADP
ijassa-348	388	30	anomaly	anomaly	NOUN
ijassa-348	388	31	intrusion	intrusion	NOUN
ijassa-348	388	32	detection	detection	NOUN
ijassa-348	388	33	”	"	PUNCT
ijassa-348	388	34	,	,	PUNCT
ijassa-348	388	35	pattern	pattern	NOUN
ijassa-348	388	36	recognition	recognition	NOUN
ijassa-348	388	37	,	,	PUNCT
ijassa-348	388	38	vol	vol	NOUN
ijassa-348	388	39	.	.	PROPN
ijassa-348	388	40	40	40	NUM
ijassa-348	388	41	,	,	PUNCT
ijassa-348	388	42	pp	pp	ADJ
ijassa-348	388	43	.	.	PUNCT
ijassa-348	388	44	2373	2373	NUM
ijassa-348	388	45	-	-	SYM
ijassa-348	388	46	2391	2391	NUM
ijassa-348	388	47	.	.	PUNCT
ijassa-348	389	1	advances	advance	NOUN
ijassa-348	389	2	in	in	ADP
ijassa-348	389	3	systems	system	NOUN
ijassa-348	389	4	science	science	NOUN
ijassa-348	389	5	and	and	CCONJ
ijassa-348	389	6	application	application	NOUN
ijassa-348	389	7	(	(	PUNCT
ijassa-348	389	8	2016	2016	NUM
ijassa-348	389	9	)	)	PUNCT
ijassa-348	389	10	vol.16	vol.16	PROPN
ijassa-348	389	11	no.2	no.2	PROPN
ijassa-348	389	12	37	37	NUM
ijassa-348	390	1	[	[	X
ijassa-348	390	2	7	7	NUM
ijassa-348	390	3	]	]	X
ijassa-348	390	4	li	li	PROPN
ijassa-348	390	5	y.	y.	PROPN
ijassa-348	390	6	and	and	CCONJ
ijassa-348	390	7	guo	guo	PROPN
ijassa-348	390	8	l.	l.	PROPN
ijassa-348	390	9	(	(	PUNCT
ijassa-348	390	10	2007	2007	NUM
ijassa-348	390	11	)	)	PUNCT
ijassa-348	390	12	,	,	PUNCT
ijassa-348	390	13	“	"	PUNCT
ijassa-348	390	14	an	an	DET
ijassa-348	390	15	active	active	ADJ
ijassa-348	390	16	learning	learning	NOUN
ijassa-348	390	17	based	base	VERB
ijassa-348	390	18	tcm	tcm	PROPN
ijassa-348	390	19	-	-	PUNCT
ijassa-348	390	20	knn	knn	PROPN
ijassa-348	390	21	algorithm	algorithm	NOUN
ijassa-348	390	22	for	for	ADP
ijassa-348	390	23	supervised	supervised	ADJ
ijassa-348	390	24	network	network	NOUN
ijassa-348	390	25	intrusion	intrusion	NOUN
ijassa-348	390	26	detection	detection	NOUN
ijassa-348	390	27	”	"	PUNCT
ijassa-348	390	28	,	,	PUNCT
ijassa-348	390	29	computer	computer	NOUN
ijassa-348	390	30	and	and	CCONJ
ijassa-348	390	31	security	security	NOUN
ijassa-348	390	32	,	,	PUNCT
ijassa-348	390	33	vol	vol	NOUN
ijassa-348	390	34	.	.	PROPN
ijassa-348	390	35	26	26	NUM
ijassa-348	390	36	,	,	PUNCT
ijassa-348	390	37	pp	pp	ADJ
ijassa-348	390	38	.	.	PUNCT
ijassa-348	391	1	459	459	NUM
ijassa-348	391	2	-	-	SYM
ijassa-348	391	3	467	467	NUM
ijassa-348	391	4	.	.	PUNCT
ijassa-348	392	1	[	[	X
ijassa-348	392	2	8	8	NUM
ijassa-348	392	3	]	]	X
ijassa-348	392	4	amor	amor	PROPN
ijassa-348	392	5	n.b	n.b	PROPN
ijassa-348	392	6	.	.	PROPN
ijassa-348	392	7	,	,	PUNCT
ijassa-348	392	8	benferhat	benferhat	ADP
ijassa-348	392	9	s.	s.	PROPN
ijassa-348	392	10	,	,	PUNCT
ijassa-348	392	11	elouedi	elouedi	PROPN
ijassa-348	392	12	z.and	z.and	PROPN
ijassa-348	392	13	zhang	zhang	PROPN
ijassa-348	392	14	x.t	x.t	PROPN
ijassa-348	392	15	.	.	PROPN
ijassa-348	393	1	(	(	PUNCT
ijassa-348	393	2	2004	2004	NUM
ijassa-348	393	3	)	)	PUNCT
ijassa-348	393	4	.	.	PUNCT
ijassa-348	394	1	näıve	näıve	NUM
ijassa-348	394	2	bayes	baye	NOUN
ijassa-348	394	3	vs	vs	ADP
ijassa-348	394	4	decision	decision	NOUN
ijassa-348	394	5	trees	tree	NOUN
ijassa-348	394	6	in	in	ADP
ijassa-348	394	7	intrusion	intrusion	NOUN
ijassa-348	394	8	detection	detection	NOUN
ijassa-348	394	9	systems	system	NOUN
ijassa-348	394	10	,	,	PUNCT
ijassa-348	394	11	sac04	sac04	ADJ
ijassa-348	394	12	:	:	PUNCT
ijassa-348	394	13	proceedings	proceeding	NOUN
ijassa-348	394	14	of	of	ADP
ijassa-348	394	15	the	the	DET
ijassa-348	394	16	2004	2004	NUM
ijassa-348	394	17	acm	acm	NOUN
ijassa-348	394	18	symposium	symposium	NOUN
ijassa-348	394	19	on	on	ADP
ijassa-348	394	20	applied	apply	VERB
ijassa-348	394	21	computing	computing	NOUN
ijassa-348	394	22	,	,	PUNCT
ijassa-348	394	23	new	new	PROPN
ijassa-348	394	24	york	york	PROPN
ijassa-348	394	25	,	,	PUNCT
ijassa-348	394	26	ny	ny	PROPN
ijassa-348	394	27	,	,	PUNCT
ijassa-348	394	28	usa	usa	PROPN
ijassa-348	394	29	:	:	PUNCT
ijassa-348	394	30	acm	acm	PROPN
ijassa-348	394	31	press	press	NOUN
ijassa-348	394	32	.	.	PUNCT
ijassa-348	395	1	[	[	X
ijassa-348	395	2	9	9	X
ijassa-348	395	3	]	]	X
ijassa-348	395	4	xiang	xiang	PROPN
ijassa-348	395	5	c.	c.	PROPN
ijassa-348	395	6	,	,	PUNCT
ijassa-348	395	7	yong	yong	PROPN
ijassa-348	395	8	p.c	p.c	PROPN
ijassa-348	395	9	.	.	PROPN
ijassa-348	395	10	and	and	CCONJ
ijassa-348	395	11	meng	meng	PROPN
ijassa-348	395	12	l.s	l.s	PROPN
ijassa-348	395	13	.	.	PUNCT
ijassa-348	396	1	(	(	PUNCT
ijassa-348	396	2	2008),“design	2008),“design	NOUN
ijassa-348	396	3	of	of	ADP
ijassa-348	396	4	multiple	multiple	ADJ
ijassa-348	396	5	-	-	PUNCT
ijassa-348	396	6	level	level	NOUN
ijassa-348	396	7	hybrid	hybrid	NOUN
ijassa-348	396	8	classifier	classifier	NOUN
ijassa-348	396	9	for	for	ADP
ijassa-348	396	10	intrusion	intrusion	NOUN
ijassa-348	396	11	detection	detection	NOUN
ijassa-348	396	12	system	system	NOUN
ijassa-348	396	13	using	use	VERB
ijassa-348	396	14	bayesian	bayesian	NOUN
ijassa-348	396	15	clustering	clustering	NOUN
ijassa-348	396	16	and	and	CCONJ
ijassa-348	396	17	decision	decision	NOUN
ijassa-348	396	18	trees	tree	NOUN
ijassa-348	396	19	”	"	PUNCT
ijassa-348	396	20	,	,	PUNCT
ijassa-348	396	21	pattern	pattern	NOUN
ijassa-348	396	22	recognition	recognition	NOUN
ijassa-348	396	23	letters	letter	NOUN
ijassa-348	396	24	,	,	PUNCT
ijassa-348	396	25	vol	vol	NOUN
ijassa-348	396	26	.	.	PROPN
ijassa-348	396	27	29	29	NUM
ijassa-348	396	28	,	,	PUNCT
ijassa-348	396	29	no	no	INTJ
ijassa-348	396	30	.	.	NOUN
ijassa-348	396	31	7	7	NUM
ijassa-348	396	32	,	,	PUNCT
ijassa-348	396	33	pp	pp	ADJ
ijassa-348	396	34	.	.	PUNCT
ijassa-348	397	1	918	918	NUM
ijassa-348	397	2	-	-	SYM
ijassa-348	397	3	924	924	NUM
ijassa-348	397	4	.	.	PUNCT
ijassa-348	398	1	[	[	X
ijassa-348	398	2	10	10	NUM
ijassa-348	398	3	]	]	X
ijassa-348	398	4	shafi	shafi	PROPN
ijassa-348	398	5	k.	k.	PROPN
ijassa-348	398	6	and	and	CCONJ
ijassa-348	398	7	abbass	abbass	PROPN
ijassa-348	398	8	h.a	h.a	PROPN
ijassa-348	398	9	.	.	PROPN
ijassa-348	398	10	(	(	PUNCT
ijassa-348	398	11	2009	2009	NUM
ijassa-348	398	12	)	)	PUNCT
ijassa-348	398	13	,	,	PUNCT
ijassa-348	398	14	“	"	PUNCT
ijassa-348	398	15	an	an	DET
ijassa-348	398	16	adaptive	adaptive	ADJ
ijassa-348	398	17	genetic	genetic	ADJ
ijassa-348	398	18	based	base	VERB
ijassa-348	398	19	signature	signature	NOUN
ijassa-348	398	20	learning	learning	NOUN
ijassa-348	398	21	system	system	NOUN
ijassa-348	398	22	for	for	ADP
ijassa-348	398	23	intrusion	intrusion	NOUN
ijassa-348	398	24	detection	detection	NOUN
ijassa-348	398	25	”	"	PUNCT
ijassa-348	398	26	,	,	PUNCT
ijassa-348	398	27	expert	expert	NOUN
ijassa-348	398	28	systems	system	NOUN
ijassa-348	398	29	with	with	ADP
ijassa-348	398	30	applications	application	NOUN
ijassa-348	398	31	,	,	PUNCT
ijassa-348	398	32	vol	vol	NOUN
ijassa-348	398	33	.	.	PROPN
ijassa-348	399	1	36	36	NUM
ijassa-348	399	2	,	,	PUNCT
ijassa-348	399	3	no	no	INTJ
ijassa-348	399	4	.	.	NOUN
ijassa-348	399	5	10	10	NUM
ijassa-348	399	6	,	,	PUNCT
ijassa-348	399	7	pp	pp	ADJ
ijassa-348	399	8	.	.	PUNCT
ijassa-348	400	1	12036	12036	NUM
ijassa-348	400	2	-	-	SYM
ijassa-348	400	3	12043	12043	NUM
ijassa-348	400	4	.	.	PUNCT
ijassa-348	401	1	[	[	X
ijassa-348	401	2	11	11	NUM
ijassa-348	401	3	]	]	X
ijassa-348	401	4	sumaiya	sumaiya	PROPN
ijassa-348	401	5	thaseen	thaseen	PROPN
ijassa-348	401	6	and	and	CCONJ
ijassa-348	401	7	ch.aswani	ch.aswani	X
ijassa-348	402	1	kumar	kumar	PROPN
ijassa-348	402	2	.	.	PUNCT
ijassa-348	402	3	(	(	PUNCT
ijassa-348	402	4	2014	2014	NUM
ijassa-348	402	5	)	)	PUNCT
ijassa-348	402	6	,	,	PUNCT
ijassa-348	402	7	“	"	PUNCT
ijassa-348	402	8	intrusion	intrusion	NOUN
ijassa-348	402	9	detection	detection	NOUN
ijassa-348	402	10	model	model	NOUN
ijassa-348	402	11	using	use	VERB
ijassa-348	402	12	fusion	fusion	NOUN
ijassa-348	402	13	of	of	ADP
ijassa-348	402	14	pca	pca	PROPN
ijassa-348	402	15	and	and	CCONJ
ijassa-348	402	16	optimized	optimize	VERB
ijassa-348	402	17	svm	svm	NOUN
ijassa-348	402	18	”	"	PUNCT
ijassa-348	402	19	,	,	PUNCT
ijassa-348	402	20	2014	2014	NUM
ijassa-348	402	21	international	international	ADJ
ijassa-348	402	22	conference	conference	NOUN
ijassa-348	402	23	on	on	ADP
ijassa-348	402	24	computing	computing	NOUN
ijassa-348	402	25	and	and	CCONJ
ijassa-348	402	26	informatics	informatic	NOUN
ijassa-348	402	27	(	(	PUNCT
ijassa-348	402	28	ic3i	ic3i	X
ijassa-348	402	29	)	)	PUNCT
ijassa-348	402	30	27	27	NUM
ijassa-348	402	31	-	-	SYM
ijassa-348	402	32	29	29	NUM
ijassa-348	402	33	nov	nov	NOUN
ijassa-348	402	34	2014	2014	NUM
ijassa-348	402	35	,	,	PUNCT
ijassa-348	402	36	pp	pp	ADV
ijassa-348	402	37	.	.	PUNCT
ijassa-348	402	38	879	879	NUM
ijassa-348	402	39	-	-	SYM
ijassa-348	402	40	884	884	NUM
ijassa-348	402	41	.	.	PUNCT
ijassa-348	403	1	[	[	X
ijassa-348	403	2	12	12	NUM
ijassa-348	403	3	]	]	PUNCT
ijassa-348	403	4	i.sumaiya	i.sumaiya	NOUN
ijassa-348	403	5	thaseen	thaseen	PROPN
ijassa-348	403	6	and	and	CCONJ
ijassa-348	403	7	ch.aswani	ch.aswani	X
ijassa-348	404	1	kumar	kumar	PROPN
ijassa-348	404	2	.	.	PUNCT
ijassa-348	405	1	(	(	PUNCT
ijassa-348	405	2	2016	2016	NUM
ijassa-348	405	3	)	)	PUNCT
ijassa-348	405	4	,	,	PUNCT
ijassa-348	405	5	“	"	PUNCT
ijassa-348	405	6	improving	improve	VERB
ijassa-348	405	7	accuracy	accuracy	NOUN
ijassa-348	405	8	of	of	ADP
ijassa-348	405	9	intrusion	intrusion	NOUN
ijassa-348	405	10	detection	detection	NOUN
ijassa-348	405	11	model	model	NOUN
ijassa-348	405	12	using	use	VERB
ijassa-348	405	13	pca	pca	PROPN
ijassa-348	405	14	and	and	CCONJ
ijassa-348	405	15	optimized	optimize	VERB
ijassa-348	405	16	svm	svm	NOUN
ijassa-348	405	17	”	"	PUNCT
ijassa-348	405	18	,	,	PUNCT
ijassa-348	405	19	cit	cit	PROPN
ijassa-348	405	20	journal	journal	PROPN
ijassa-348	405	21	of	of	ADP
ijassa-348	405	22	computing	computing	NOUN
ijassa-348	405	23	and	and	CCONJ
ijassa-348	405	24	information	information	NOUN
ijassa-348	405	25	technology	technology	NOUN
ijassa-348	405	26	,	,	PUNCT
ijassa-348	405	27	vol.24	vol.24	NOUN
ijassa-348	405	28	,	,	PUNCT
ijassa-348	405	29	no.2	no.2	PROPN
ijassa-348	405	30	,	,	PUNCT
ijassa-348	405	31	pp	pp	ADJ
ijassa-348	405	32	.	.	PUNCT
ijassa-348	406	1	133	133	NUM
ijassa-348	406	2	-	-	SYM
ijassa-348	406	3	148	148	NUM
ijassa-348	406	4	[	[	SYM
ijassa-348	406	5	13	13	NUM
ijassa-348	406	6	]	]	X
ijassa-348	406	7	ajith	ajith	PROPN
ijassa-348	406	8	abraham	abraham	PROPN
ijassa-348	406	9	,	,	PUNCT
ijassa-348	406	10	crina	crina	PROPN
ijassa-348	406	11	grosan	grosan	PROPN
ijassa-348	406	12	and	and	CCONJ
ijassa-348	406	13	carlos	carlos	PROPN
ijassa-348	406	14	martin	martin	PROPN
ijassa-348	406	15	-	-	PUNCT
ijassa-348	406	16	vide	vide	NOUN
ijassa-348	406	17	.	.	PUNCT
ijassa-348	407	1	(	(	PUNCT
ijassa-348	407	2	2007	2007	NUM
ijassa-348	407	3	)	)	PUNCT
ijassa-348	407	4	,	,	PUNCT
ijassa-348	407	5	“	"	PUNCT
ijassa-348	407	6	evolutionary	evolutionary	ADJ
ijassa-348	407	7	design	design	NOUN
ijassa-348	407	8	of	of	ADP
ijassa-348	407	9	intrusion	intrusion	NOUN
ijassa-348	407	10	detection	detection	NOUN
ijassa-348	407	11	programs	program	NOUN
ijassa-348	407	12	”	"	PUNCT
ijassa-348	407	13	,	,	PUNCT
ijassa-348	407	14	international	international	ADJ
ijassa-348	407	15	journal	journal	NOUN
ijassa-348	407	16	of	of	ADP
ijassa-348	407	17	network	network	NOUN
ijassa-348	407	18	security	security	NOUN
ijassa-348	407	19	,	,	PUNCT
ijassa-348	407	20	vol	vol	NOUN
ijassa-348	407	21	.	.	PROPN
ijassa-348	407	22	4	4	NUM
ijassa-348	407	23	,	,	PUNCT
ijassa-348	407	24	no	no	INTJ
ijassa-348	407	25	.	.	NOUN
ijassa-348	407	26	3	3	NUM
ijassa-348	407	27	,	,	PUNCT
ijassa-348	407	28	pp	pp	ADJ
ijassa-348	407	29	.	.	PUNCT
ijassa-348	408	1	328	328	NUM
ijassa-348	408	2	-	-	SYM
ijassa-348	408	3	339	339	NUM
ijassa-348	408	4	.	.	PUNCT
ijassa-348	409	1	[	[	X
ijassa-348	409	2	14	14	NUM
ijassa-348	409	3	]	]	X
ijassa-348	409	4	fangjun	fangjun	PROPN
ijassa-348	409	5	kuanga	kuanga	NOUN
ijassa-348	409	6	,	,	PUNCT
ijassa-348	409	7	weihong	weihong	ADJ
ijassa-348	409	8	xua	xua	PROPN
ijassa-348	409	9	and	and	CCONJ
ijassa-348	409	10	siyang	siyang	PROPN
ijassa-348	409	11	zhang.(2014	zhang.(2014	NUM
ijassa-348	409	12	)	)	PUNCT
ijassa-348	409	13	.	.	PUNCT
ijassa-348	410	1	“	"	PUNCT
ijassa-348	410	2	a	a	DET
ijassa-348	410	3	novel	novel	ADJ
ijassa-348	410	4	hybrid	hybrid	ADJ
ijassa-348	410	5	kpca	kpca	NOUN
ijassa-348	410	6	and	and	CCONJ
ijassa-348	410	7	svm	svm	VERB
ijassa-348	410	8	with	with	ADP
ijassa-348	410	9	ga	ga	PROPN
ijassa-348	410	10	model	model	NOUN
ijassa-348	410	11	for	for	ADP
ijassa-348	410	12	intrusion	intrusion	NOUN
ijassa-348	410	13	detection	detection	NOUN
ijassa-348	410	14	”	"	PUNCT
ijassa-348	410	15	,	,	PUNCT
ijassa-348	410	16	in	in	ADP
ijassa-348	410	17	applied	apply	VERB
ijassa-348	410	18	soft	soft	ADJ
ijassa-348	410	19	computing	computing	NOUN
ijassa-348	410	20	,	,	PUNCT
ijassa-348	410	21	vol	vol	NOUN
ijassa-348	410	22	.	.	PROPN
ijassa-348	410	23	18	18	NUM
ijassa-348	410	24	,	,	PUNCT
ijassa-348	410	25	pp	pp	ADJ
ijassa-348	410	26	.	.	PUNCT
ijassa-348	411	1	178	178	NUM
ijassa-348	411	2	-	-	SYM
ijassa-348	411	3	184	184	NUM
ijassa-348	411	4	.	.	PUNCT
ijassa-348	412	1	[	[	X
ijassa-348	412	2	15	15	NUM
ijassa-348	412	3	]	]	X
ijassa-348	412	4	srilatha	srilatha	NOUN
ijassa-348	412	5	chebrolu	chebrolu	PROPN
ijassa-348	412	6	,	,	PUNCT
ijassa-348	412	7	ajith	ajith	PROPN
ijassa-348	412	8	abraham	abraham	PROPN
ijassa-348	412	9	and	and	CCONJ
ijassa-348	412	10	johnson	johnson	PROPN
ijassa-348	412	11	p.thomas	p.thomas	PROPN
ijassa-348	412	12	.	.	PUNCT
ijassa-348	413	1	(	(	PUNCT
ijassa-348	413	2	2005	2005	NUM
ijassa-348	413	3	)	)	PUNCT
ijassa-348	413	4	,	,	PUNCT
ijassa-348	413	5	“	"	PUNCT
ijassa-348	413	6	feature	feature	NOUN
ijassa-348	413	7	deduction	deduction	NOUN
ijassa-348	413	8	and	and	CCONJ
ijassa-348	413	9	ensemble	ensemble	ADJ
ijassa-348	413	10	design	design	NOUN
ijassa-348	413	11	of	of	ADP
ijassa-348	413	12	intrusion	intrusion	NOUN
ijassa-348	413	13	detection	detection	NOUN
ijassa-348	413	14	systems	system	NOUN
ijassa-348	413	15	”	"	PUNCT
ijassa-348	413	16	,	,	PUNCT
ijassa-348	413	17	computers	computer	NOUN
ijassa-348	413	18	and	and	CCONJ
ijassa-348	413	19	security	security	NOUN
ijassa-348	413	20	,	,	PUNCT
ijassa-348	413	21	vol	vol	NOUN
ijassa-348	413	22	.	.	PROPN
ijassa-348	414	1	24	24	NUM
ijassa-348	414	2	,	,	PUNCT
ijassa-348	414	3	pp	pp	ADJ
ijassa-348	414	4	.	.	PUNCT
ijassa-348	415	1	295	295	NUM
ijassa-348	415	2	-	-	SYM
ijassa-348	415	3	307	307	NUM
ijassa-348	415	4	.	.	PUNCT
ijassa-348	416	1	[	[	X
ijassa-348	416	2	16	16	NUM
ijassa-348	416	3	]	]	X
ijassa-348	416	4	alexandre	alexandre	PROPN
ijassa-348	416	5	balon	balon	PROPN
ijassa-348	416	6	-	-	PUNCT
ijassa-348	416	7	perin	perin	NOUN
ijassa-348	416	8	.	.	PUNCT
ijassa-348	417	1	(	(	PUNCT
ijassa-348	417	2	2012	2012	NUM
ijassa-348	417	3	)	)	PUNCT
ijassa-348	417	4	,	,	PUNCT
ijassa-348	417	5	“	"	PUNCT
ijassa-348	417	6	ensemble	ensemble	ADJ
ijassa-348	417	7	-	-	PUNCT
ijassa-348	417	8	based	base	VERB
ijassa-348	417	9	methods	method	NOUN
ijassa-348	417	10	for	for	ADP
ijassa-348	417	11	intrusion	intrusion	NOUN
ijassa-348	417	12	detection	detection	NOUN
ijassa-348	417	13	”	"	PUNCT
ijassa-348	417	14	,	,	PUNCT
ijassa-348	417	15	norwegian	norwegian	ADJ
ijassa-348	417	16	university	university	NOUN
ijassa-348	417	17	of	of	ADP
ijassa-348	417	18	science	science	NOUN
ijassa-348	417	19	and	and	CCONJ
ijassa-348	417	20	technology	technology	NOUN
ijassa-348	417	21	.	.	PUNCT
ijassa-348	418	1	[	[	X
ijassa-348	418	2	17	17	NUM
ijassa-348	418	3	]	]	X
ijassa-348	418	4	chandra	chandra	PROPN
ijassa-348	418	5	a.	a.	PROPN
ijassa-348	418	6	and	and	CCONJ
ijassa-348	418	7	yao	yao	PROPN
ijassa-348	418	8	x.(2006	x.(2006	PROPN
ijassa-348	418	9	)	)	PUNCT
ijassa-348	418	10	,	,	PUNCT
ijassa-348	418	11	“	"	PUNCT
ijassa-348	418	12	evolving	evolve	VERB
ijassa-348	418	13	hybrid	hybrid	ADJ
ijassa-348	418	14	ensembles	ensemble	NOUN
ijassa-348	418	15	of	of	ADP
ijassa-348	418	16	learning	learn	VERB
ijassa-348	418	17	machines	machine	NOUN
ijassa-348	418	18	for	for	ADP
ijassa-348	418	19	better	well	ADJ
ijassa-348	418	20	generalization	generalization	NOUN
ijassa-348	418	21	”	"	PUNCT
ijassa-348	418	22	,	,	PUNCT
ijassa-348	418	23	neurocomputing	neurocomputing	NOUN
ijassa-348	418	24	,	,	PUNCT
ijassa-348	418	25	vol	vol	NOUN
ijassa-348	418	26	.	.	PROPN
ijassa-348	419	1	69	69	NUM
ijassa-348	419	2	,	,	PUNCT
ijassa-348	419	3	no	no	INTJ
ijassa-348	419	4	.	.	NOUN
ijassa-348	419	5	7	7	NUM
ijassa-348	419	6	,	,	PUNCT
ijassa-348	419	7	pp	pp	ADJ
ijassa-348	419	8	.	.	PUNCT
ijassa-348	420	1	686700	686700	NUM
ijassa-348	420	2	.	.	PUNCT
ijassa-348	421	1	38	38	NUM
ijassa-348	421	2	sumaiya	sumaiya	NOUN
ijassa-348	421	3	thaseen	thaseen	PROPN
ijassa-348	421	4	and	and	CCONJ
ijassa-348	421	5	ch.aswani	ch.aswani	X
ijassa-348	422	1	kumar	kumar	PROPN
ijassa-348	422	2	:	:	PUNCT
ijassa-348	422	3	intrusion	intrusion	NOUN
ijassa-348	422	4	detection	detection	NOUN
ijassa-348	422	5	model	model	NOUN
ijassa-348	422	6	using	use	VERB
ijassa-348	422	7	pca	pca	PROPN
ijassa-348	422	8	and	and	CCONJ
ijassa-348	422	9	...	...	PUNCT
ijassa-348	423	1	[	[	X
ijassa-348	423	2	18	18	NUM
ijassa-348	423	3	]	]	PUNCT
ijassa-348	423	4	srinivas	srinivas	PROPN
ijassa-348	423	5	mukkamala	mukkamala	PROPN
ijassa-348	423	6	,	,	PUNCT
ijassa-348	423	7	andrew	andrew	PROPN
ijassa-348	423	8	h.	h.	PROPN
ijassa-348	423	9	sung	sung	PROPN
ijassa-348	423	10	,	,	PUNCT
ijassa-348	423	11	ajith	ajith	PROPN
ijassa-348	423	12	abraham	abraham	PROPN
ijassa-348	423	13	.	.	PUNCT
ijassa-348	424	1	(	(	PUNCT
ijassa-348	424	2	2007	2007	NUM
ijassa-348	424	3	)	)	PUNCT
ijassa-348	424	4	.	.	PUNCT
ijassa-348	425	1	“	"	PUNCT
ijassa-348	425	2	intrusion	intrusion	NOUN
ijassa-348	425	3	detection	detection	NOUN
ijassa-348	425	4	using	use	VERB
ijassa-348	425	5	and	and	CCONJ
ijassa-348	425	6	ensemble	ensemble	ADJ
ijassa-348	425	7	of	of	ADP
ijassa-348	425	8	intelligent	intelligent	ADJ
ijassa-348	425	9	paradigms	paradigm	NOUN
ijassa-348	425	10	”	"	PUNCT
ijassa-348	425	11	,	,	PUNCT
ijassa-348	425	12	journal	journal	NOUN
ijassa-348	425	13	of	of	ADP
ijassa-348	425	14	network	network	NOUN
ijassa-348	425	15	and	and	CCONJ
ijassa-348	425	16	computer	computer	NOUN
ijassa-348	425	17	applications	application	NOUN
ijassa-348	425	18	,	,	PUNCT
ijassa-348	425	19	vol	vol	NOUN
ijassa-348	425	20	.	.	PROPN
ijassa-348	425	21	28	28	NUM
ijassa-348	425	22	,	,	PUNCT
ijassa-348	425	23	no	no	INTJ
ijassa-348	425	24	.	.	NOUN
ijassa-348	425	25	2	2	NUM
ijassa-348	425	26	,	,	PUNCT
ijassa-348	425	27	pp	pp	ADJ
ijassa-348	425	28	.	.	PUNCT
ijassa-348	426	1	167	167	NUM
ijassa-348	426	2	-	-	SYM
ijassa-348	426	3	182	182	NUM
ijassa-348	426	4	.	.	PUNCT
ijassa-348	427	1	[	[	X
ijassa-348	427	2	19	19	NUM
ijassa-348	427	3	]	]	SYM
ijassa-348	427	4	i.syarif	i.syarif	PROPN
ijassa-348	427	5	,	,	PUNCT
ijassa-348	427	6	e.zaluska	e.zaluska	NOUN
ijassa-348	427	7	,	,	PUNCT
ijassa-348	427	8	a.prugel	a.prugel	NOUN
ijassa-348	427	9	-	-	PUNCT
ijassa-348	427	10	bennett	bennett	NOUN
ijassa-348	427	11	and	and	CCONJ
ijassa-348	427	12	g.wills	g.wills	PROPN
ijassa-348	427	13	.	.	PUNCT
ijassa-348	428	1	(	(	PUNCT
ijassa-348	428	2	2012	2012	NUM
ijassa-348	428	3	)	)	PUNCT
ijassa-348	428	4	,	,	PUNCT
ijassa-348	428	5	“	"	PUNCT
ijassa-348	428	6	application	application	NOUN
ijassa-348	428	7	of	of	ADP
ijassa-348	428	8	bagging	bagging	NOUN
ijassa-348	428	9	,	,	PUNCT
ijassa-348	428	10	boosting	boost	VERB
ijassa-348	428	11	and	and	CCONJ
ijassa-348	428	12	stacking	stack	VERB
ijassa-348	428	13	to	to	ADP
ijassa-348	428	14	intrusion	intrusion	NOUN
ijassa-348	428	15	detection	detection	NOUN
ijassa-348	428	16	”	"	PUNCT
ijassa-348	428	17	,	,	PUNCT
ijassa-348	428	18	in	in	ADP
ijassa-348	428	19	:	:	PUNCT
ijassa-348	428	20	machine	machine	NOUN
ijassa-348	428	21	learning	learning	NOUN
ijassa-348	428	22	and	and	CCONJ
ijassa-348	428	23	data	datum	NOUN
ijassa-348	428	24	mining	mining	NOUN
ijassa-348	428	25	in	in	ADP
ijassa-348	428	26	pattern	pattern	NOUN
ijassa-348	428	27	recognition	recognition	NOUN
ijassa-348	428	28	,	,	PUNCT
ijassa-348	428	29	pp	pp	PROPN
ijassa-348	428	30	.	.	PUNCT
ijassa-348	429	1	593	593	NUM
ijassa-348	429	2	-	-	SYM
ijassa-348	429	3	602	602	NUM
ijassa-348	429	4	.	.	PUNCT
ijassa-348	430	1	[	[	X
ijassa-348	430	2	20	20	NUM
ijassa-348	430	3	]	]	X
ijassa-348	430	4	e.	e.	PROPN
ijassa-348	430	5	bahri	bahri	PROPN
ijassa-348	430	6	,	,	PUNCT
ijassa-348	430	7	n.harbi	n.harbi	NOUN
ijassa-348	430	8	and	and	CCONJ
ijassa-348	430	9	h.n.huu	h.n.huu	PROPN
ijassa-348	430	10	.	.	PUNCT
ijassa-348	431	1	(	(	PUNCT
ijassa-348	431	2	2011	2011	NUM
ijassa-348	431	3	)	)	PUNCT
ijassa-348	431	4	,	,	PUNCT
ijassa-348	431	5	“	"	PUNCT
ijassa-348	431	6	approach	approach	NOUN
ijassa-348	431	7	based	base	VERB
ijassa-348	431	8	ensemble	ensemble	ADJ
ijassa-348	431	9	methods	method	NOUN
ijassa-348	431	10	for	for	ADP
ijassa-348	431	11	better	well	ADJ
ijassa-348	431	12	and	and	CCONJ
ijassa-348	431	13	faster	fast	ADJ
ijassa-348	431	14	intrusion	intrusion	NOUN
ijassa-348	431	15	detection	detection	NOUN
ijassa-348	431	16	”	"	PUNCT
ijassa-348	431	17	,	,	PUNCT
ijassa-348	431	18	in	in	ADP
ijassa-348	431	19	:	:	PUNCT
ijassa-348	431	20	computational	computational	ADJ
ijassa-348	431	21	intelligence	intelligence	NOUN
ijassa-348	431	22	in	in	ADP
ijassa-348	431	23	security	security	NOUN
ijassa-348	431	24	for	for	ADP
ijassa-348	431	25	information	information	NOUN
ijassa-348	431	26	systems	system	NOUN
ijassa-348	431	27	,	,	PUNCT
ijassa-348	431	28	pp	pp	ADJ
ijassa-348	431	29	.	.	PUNCT
ijassa-348	432	1	17	17	NUM
ijassa-348	432	2	-	-	SYM
ijassa-348	432	3	24	24	NUM
ijassa-348	432	4	.	.	PUNCT
ijassa-348	433	1	[	[	X
ijassa-348	433	2	21	21	NUM
ijassa-348	433	3	]	]	SYM
ijassa-348	433	4	v.bukhtoyarov	v.bukhtoyarov	NOUN
ijassa-348	433	5	and	and	CCONJ
ijassa-348	433	6	v.zhukov	v.zhukov	NOUN
ijassa-348	433	7	.	.	PUNCT
ijassa-348	434	1	(	(	PUNCT
ijassa-348	434	2	2014	2014	NUM
ijassa-348	434	3	)	)	PUNCT
ijassa-348	434	4	,	,	PUNCT
ijassa-348	434	5	“	"	PUNCT
ijassa-348	434	6	ensemble	ensemble	ADJ
ijassa-348	434	7	-	-	PUNCT
ijassa-348	434	8	distributed	distribute	VERB
ijassa-348	434	9	approach	approach	NOUN
ijassa-348	434	10	in	in	ADP
ijassa-348	434	11	classification	classification	NOUN
ijassa-348	434	12	problem	problem	NOUN
ijassa-348	434	13	solution	solution	NOUN
ijassa-348	434	14	for	for	ADP
ijassa-348	434	15	intrusion	intrusion	NOUN
ijassa-348	434	16	detection	detection	NOUN
ijassa-348	434	17	systems	system	NOUN
ijassa-348	434	18	”	"	PUNCT
ijassa-348	434	19	,	,	PUNCT
ijassa-348	434	20	in	in	ADP
ijassa-348	434	21	:	:	PUNCT
ijassa-348	434	22	intelligent	intelligent	ADJ
ijassa-348	434	23	data	datum	NOUN
ijassa-348	434	24	engineering	engineering	NOUN
ijassa-348	434	25	and	and	CCONJ
ijassa-348	434	26	automated	automated	ADJ
ijassa-348	434	27	learning	learning	NOUN
ijassa-348	434	28	-	-	PUNCT
ijassa-348	434	29	ideal	ideal	NOUN
ijassa-348	434	30	2014	2014	NUM
ijassa-348	434	31	,	,	PUNCT
ijassa-348	434	32	pp	pp	ADJ
ijassa-348	434	33	.	.	PUNCT
ijassa-348	435	1	662	662	NUM
ijassa-348	435	2	-	-	SYM
ijassa-348	435	3	668	668	NUM
ijassa-348	435	4	.	.	PUNCT
ijassa-348	436	1	[	[	X
ijassa-348	436	2	22	22	NUM
ijassa-348	436	3	]	]	X
ijassa-348	436	4	cordeiro	cordeiro	PROPN
ijassa-348	436	5	junior	junior	PROPN
ijassa-348	436	6	,	,	PUNCT
ijassa-348	436	7	z.	z.	PROPN
ijassa-348	436	8	and	and	CCONJ
ijassa-348	436	9	g.l.pappa	g.l.pappa	PROPN
ijassa-348	436	10	.	.	PUNCT
ijassa-348	436	11	(	(	PUNCT
ijassa-348	436	12	2011	2011	NUM
ijassa-348	436	13	)	)	PUNCT
ijassa-348	436	14	,	,	PUNCT
ijassa-348	436	15	“	"	PUNCT
ijassa-348	436	16	a	a	DET
ijassa-348	436	17	pso	pso	NOUN
ijassa-348	436	18	algorithm	algorithm	NOUN
ijassa-348	436	19	for	for	ADP
ijassa-348	436	20	improving	improve	VERB
ijassa-348	436	21	multi	multi	ADJ
ijassa-348	436	22	-	-	ADJ
ijassa-348	436	23	view	view	ADJ
ijassa-348	436	24	classification	classification	NOUN
ijassa-348	436	25	”	"	PUNCT
ijassa-348	436	26	,	,	PUNCT
ijassa-348	436	27	in	in	ADP
ijassa-348	436	28	:	:	PUNCT
ijassa-348	436	29	2011	2011	NUM
ijassa-348	436	30	ieee	ieee	PROPN
ijassa-348	436	31	congress	congress	PROPN
ijassa-348	436	32	on	on	ADP
ijassa-348	436	33	evolutionary	evolutionary	ADJ
ijassa-348	436	34	computation	computation	NOUN
ijassa-348	436	35	(	(	PUNCT
ijassa-348	436	36	cec	cec	PROPN
ijassa-348	436	37	)	)	PUNCT
ijassa-348	436	38	,	,	PUNCT
ijassa-348	436	39	pp	pp	PROPN
ijassa-348	436	40	.	.	PUNCT
ijassa-348	437	1	925	925	NUM
ijassa-348	437	2	-	-	SYM
ijassa-348	437	3	932	932	NUM
ijassa-348	437	4	.	.	PUNCT
ijassa-348	438	1	[	[	X
ijassa-348	438	2	23	23	NUM
ijassa-348	438	3	]	]	PUNCT
ijassa-348	438	4	a.kausar	a.kausar	VERB
ijassa-348	438	5	,	,	PUNCT
ijassa-348	438	6	m.ishtiaq	m.ishtiaq	ADJ
ijassa-348	438	7	,	,	PUNCT
ijassa-348	438	8	m.a.jaffar	m.a.jaffar	ADJ
ijassa-348	438	9	and	and	CCONJ
ijassa-348	438	10	a.m.mirza	a.m.mirza	ADJ
ijassa-348	438	11	.	.	PUNCT
ijassa-348	439	1	(	(	PUNCT
ijassa-348	439	2	2010	2010	NUM
ijassa-348	439	3	)	)	PUNCT
ijassa-348	439	4	,	,	PUNCT
ijassa-348	439	5	“	"	PUNCT
ijassa-348	439	6	optimization	optimization	NOUN
ijassa-348	439	7	of	of	ADP
ijassa-348	439	8	ensemble	ensemble	ADJ
ijassa-348	439	9	based	base	VERB
ijassa-348	439	10	decision	decision	NOUN
ijassa-348	439	11	using	use	VERB
ijassa-348	439	12	pco	pco	NOUN
ijassa-348	439	13	”	"	PUNCT
ijassa-348	439	14	,	,	PUNCT
ijassa-348	439	15	in	in	ADP
ijassa-348	439	16	proceedings	proceeding	NOUN
ijassa-348	439	17	of	of	ADP
ijassa-348	439	18	the	the	DET
ijassa-348	439	19	world	world	NOUN
ijassa-348	439	20	congress	congress	PROPN
ijassa-348	439	21	on	on	ADP
ijassa-348	439	22	engineering	engineering	PROPN
ijassa-348	439	23	,	,	PUNCT
ijassa-348	439	24	wce	wce	X
ijassa-348	439	25	,	,	PUNCT
ijassa-348	439	26	vol	vol	NOUN
ijassa-348	439	27	.	.	PROPN
ijassa-348	439	28	10	10	NUM
ijassa-348	439	29	.	.	PUNCT
ijassa-348	440	1	[	[	X
ijassa-348	440	2	24	24	NUM
ijassa-348	440	3	]	]	PUNCT
ijassa-348	440	4	http://nsl.cs.unb.ca/nsl-kdd/	http://nsl.cs.unb.ca/nsl-kdd/	PROPN
ijassa-348	440	5	[	[	X
ijassa-348	440	6	25	25	NUM
ijassa-348	440	7	]	]	PUNCT
ijassa-348	440	8	s.j.horng	s.j.horng	NOUN
ijassa-348	440	9	,	,	PUNCT
ijassa-348	440	10	m.y.su	m.y.su	PROPN
ijassa-348	440	11	,	,	PUNCT
ijassa-348	440	12	y.h.chen	y.h.chen	ADV
ijassa-348	440	13	,	,	PUNCT
ijassa-348	440	14	t.w.kao	t.w.kao	PROPN
ijassa-348	440	15	,	,	PUNCT
ijassa-348	440	16	r.j.chen	r.j.chen	NOUN
ijassa-348	440	17	,	,	PUNCT
ijassa-348	440	18	j.l.lai	j.l.lai	PROPN
ijassa-348	440	19	and	and	CCONJ
ijassa-348	440	20	c.d.perkasa	c.d.perkasa	PROPN
ijassa-348	440	21	.	.	PUNCT
ijassa-348	441	1	(	(	PUNCT
ijassa-348	441	2	2011	2011	NUM
ijassa-348	441	3	)	)	PUNCT
ijassa-348	441	4	,	,	PUNCT
ijassa-348	441	5	“	"	PUNCT
ijassa-348	441	6	a	a	DET
ijassa-348	441	7	novel	novel	ADJ
ijassa-348	441	8	intrusion	intrusion	NOUN
ijassa-348	441	9	detection	detection	NOUN
ijassa-348	441	10	system	system	NOUN
ijassa-348	441	11	based	base	VERB
ijassa-348	441	12	on	on	ADP
ijassa-348	441	13	hierarchical	hierarchical	ADJ
ijassa-348	441	14	clustering	clustering	NOUN
ijassa-348	441	15	and	and	CCONJ
ijassa-348	441	16	support	support	NOUN
ijassa-348	441	17	vector	vector	NOUN
ijassa-348	441	18	machines	machine	NOUN
ijassa-348	441	19	”	"	PUNCT
ijassa-348	441	20	,	,	PUNCT
ijassa-348	441	21	expert	expert	NOUN
ijassa-348	441	22	systems	system	NOUN
ijassa-348	441	23	and	and	CCONJ
ijassa-348	441	24	applications	application	NOUN
ijassa-348	441	25	,	,	PUNCT
ijassa-348	441	26	vol	vol	NOUN
ijassa-348	441	27	.	.	PROPN
ijassa-348	442	1	38	38	NUM
ijassa-348	442	2	,	,	PUNCT
ijassa-348	442	3	no	no	INTJ
ijassa-348	442	4	.	.	NOUN
ijassa-348	442	5	1	1	NUM
ijassa-348	442	6	,	,	PUNCT
ijassa-348	442	7	pp	pp	ADJ
ijassa-348	442	8	.	.	PUNCT
ijassa-348	443	1	306	306	NUM
ijassa-348	443	2	-	-	SYM
ijassa-348	443	3	313	313	NUM
ijassa-348	443	4	.	.	PUNCT
ijassa-348	444	1	corresponding	correspond	VERB
ijassa-348	444	2	author	author	NOUN
ijassa-348	444	3	sumaiya	sumaiya	PROPN
ijassa-348	444	4	thaseen	thaseen	PROPN
ijassa-348	444	5	can	can	AUX
ijassa-348	444	6	be	be	AUX
ijassa-348	444	7	contacted	contact	VERB
ijassa-348	444	8	at	at	ADP
ijassa-348	444	9	:	:	PUNCT
ijassa-348	444	10	sumaiyathaseen@gmail.com	sumaiyathaseen@gmail.com	X