id	sid	tid	token	lemma	pos
fcis-10369	1	1	frontiers	frontier	NOUN
fcis-10369	1	2	in	in	ADP
fcis-10369	1	3	computing	computing	NOUN
fcis-10369	1	4	and	and	CCONJ
fcis-10369	1	5	intelligent	intelligent	ADJ
fcis-10369	1	6	systems	system	NOUN
fcis-10369	1	7	issn	issn	VERB
fcis-10369	1	8	:	:	PUNCT
fcis-10369	1	9	2832	2832	NUM
fcis-10369	1	10	-	-	SYM
fcis-10369	1	11	6024	6024	NUM
fcis-10369	1	12	|	|	NOUN
fcis-10369	1	13	vol	vol	NOUN
fcis-10369	1	14	.	.	PROPN
fcis-10369	2	1	4	4	NUM
fcis-10369	2	2	,	,	PUNCT
fcis-10369	2	3	no	no	INTJ
fcis-10369	2	4	.	.	NOUN
fcis-10369	2	5	2	2	NUM
fcis-10369	2	6	,	,	PUNCT
fcis-10369	2	7	2023	2023	NUM
fcis-10369	2	8	124	124	NUM
fcis-10369	2	9	an	an	DET
fcis-10369	2	10	intrusion	intrusion	NOUN
fcis-10369	2	11	detection	detection	NOUN
fcis-10369	2	12	method	method	NOUN
fcis-10369	2	13	based	base	VERB
fcis-10369	2	14	on	on	ADP
fcis-10369	2	15	fusion	fusion	NOUN
fcis-10369	2	16	neural	neural	ADJ
fcis-10369	2	17	network	network	NOUN
fcis-10369	2	18	xin	xin	PROPN
fcis-10369	2	19	li	li	PROPN
fcis-10369	2	20	1	1	NUM
fcis-10369	2	21	,	,	PUNCT
fcis-10369	2	22	hong	hong	PROPN
fcis-10369	2	23	huang	huang	PROPN
fcis-10369	2	24	1	1	NUM
fcis-10369	2	25	,	,	PUNCT
fcis-10369	2	26	2	2	NUM
fcis-10369	2	27	,	,	PUNCT
fcis-10369	2	28	guotao	guotao	ADJ
fcis-10369	2	29	yuan	yuan	PROPN
fcis-10369	2	30	1	1	NUM
fcis-10369	2	31	,	,	PUNCT
fcis-10369	2	32	zhaolian	zhaolian	NOUN
fcis-10369	2	33	wang	wang	PROPN
fcis-10369	2	34	1	1	NUM
fcis-10369	2	35	,	,	PUNCT
fcis-10369	2	36	rui	rui	PROPN
fcis-10369	2	37	du	du	PROPN
fcis-10369	2	38	1	1	NUM
fcis-10369	2	39	1	1	NUM
fcis-10369	2	40	school	school	NOUN
fcis-10369	2	41	of	of	ADP
fcis-10369	2	42	computer	computer	NOUN
fcis-10369	2	43	science	science	NOUN
fcis-10369	2	44	and	and	CCONJ
fcis-10369	2	45	engineering	engineering	NOUN
fcis-10369	2	46	,	,	PUNCT
fcis-10369	2	47	sichuan	sichuan	PROPN
fcis-10369	2	48	university	university	PROPN
fcis-10369	2	49	of	of	ADP
fcis-10369	2	50	science	science	PROPN
fcis-10369	2	51	&	&	CCONJ
fcis-10369	2	52	engineering	engineering	PROPN
fcis-10369	2	53	,	,	PUNCT
fcis-10369	2	54	yibin	yibin	PROPN
fcis-10369	2	55	644000	644000	NUM
fcis-10369	2	56	,	,	PUNCT
fcis-10369	2	57	china	china	PROPN
fcis-10369	2	58	2	2	NUM
fcis-10369	2	59	key	key	ADJ
fcis-10369	2	60	laboratory	laboratory	NOUN
fcis-10369	2	61	of	of	ADP
fcis-10369	2	62	higher	high	ADJ
fcis-10369	2	63	education	education	NOUN
fcis-10369	2	64	of	of	ADP
fcis-10369	2	65	sichuan	sichuan	PROPN
fcis-10369	2	66	province	province	PROPN
fcis-10369	2	67	for	for	ADP
fcis-10369	2	68	enterprise	enterprise	NOUN
fcis-10369	2	69	informationalization	informationalization	NOUN
fcis-10369	2	70	and	and	CCONJ
fcis-10369	2	71	internet	internet	NOUN
fcis-10369	2	72	of	of	ADP
fcis-10369	2	73	things	thing	NOUN
fcis-10369	2	74	,	,	PUNCT
fcis-10369	2	75	yibin	yibin	PROPN
fcis-10369	2	76	644000	644000	NUM
fcis-10369	2	77	,	,	PUNCT
fcis-10369	2	78	china	china	PROPN
fcis-10369	2	79	abstract	abstract	NOUN
fcis-10369	2	80	:	:	PUNCT
fcis-10369	2	81	aiming	aim	VERB
fcis-10369	2	82	at	at	ADP
fcis-10369	2	83	the	the	DET
fcis-10369	2	84	problems	problem	NOUN
fcis-10369	2	85	of	of	ADP
fcis-10369	2	86	class	class	NOUN
fcis-10369	2	87	imbalance	imbalance	NOUN
fcis-10369	2	88	,	,	PUNCT
fcis-10369	2	89	insufficient	insufficient	ADJ
fcis-10369	2	90	feature	feature	NOUN
fcis-10369	2	91	learning	learning	NOUN
fcis-10369	2	92	,	,	PUNCT
fcis-10369	2	93	weak	weak	ADJ
fcis-10369	2	94	generalization	generalization	NOUN
fcis-10369	2	95	ability	ability	NOUN
fcis-10369	2	96	,	,	PUNCT
fcis-10369	2	97	and	and	CCONJ
fcis-10369	2	98	representation	representation	NOUN
fcis-10369	2	99	capability	capability	NOUN
fcis-10369	2	100	in	in	ADP
fcis-10369	2	101	existing	exist	VERB
fcis-10369	2	102	intrusion	intrusion	NOUN
fcis-10369	2	103	detection	detection	NOUN
fcis-10369	2	104	models	model	NOUN
fcis-10369	2	105	,	,	PUNCT
fcis-10369	2	106	we	we	PRON
fcis-10369	2	107	propose	propose	VERB
fcis-10369	2	108	a	a	DET
fcis-10369	2	109	multi	multi	ADJ
fcis-10369	2	110	-	-	ADJ
fcis-10369	2	111	scale	scale	ADJ
fcis-10369	2	112	feature	feature	NOUN
fcis-10369	2	113	fusion	fusion	NOUN
fcis-10369	2	114	intrusion	intrusion	NOUN
fcis-10369	2	115	detection	detection	NOUN
fcis-10369	2	116	model	model	NOUN
fcis-10369	2	117	(	(	PUNCT
fcis-10369	2	118	msff	msff	NOUN
fcis-10369	2	119	)	)	PUNCT
fcis-10369	2	120	.	.	PUNCT
fcis-10369	3	1	this	this	DET
fcis-10369	3	2	model	model	NOUN
fcis-10369	3	3	combines	combine	VERB
fcis-10369	3	4	multi	multi	ADJ
fcis-10369	3	5	-	-	ADJ
fcis-10369	3	6	scale	scale	ADJ
fcis-10369	3	7	one	one	NUM
fcis-10369	3	8	-	-	PUNCT
fcis-10369	3	9	dimensional	dimensional	ADJ
fcis-10369	3	10	convolution	convolution	NOUN
fcis-10369	3	11	and	and	CCONJ
fcis-10369	3	12	bidirectional	bidirectional	ADJ
fcis-10369	3	13	long	long	ADJ
fcis-10369	3	14	short	short	ADJ
fcis-10369	3	15	-	-	PUNCT
fcis-10369	3	16	term	term	NOUN
fcis-10369	3	17	memory	memory	NOUN
fcis-10369	3	18	(	(	PUNCT
fcis-10369	3	19	lstm	lstm	NOUN
fcis-10369	3	20	)	)	PUNCT
fcis-10369	3	21	networks	network	NOUN
fcis-10369	3	22	,	,	PUNCT
fcis-10369	3	23	and	and	CCONJ
fcis-10369	3	24	incorporates	incorporate	VERB
fcis-10369	3	25	residual	residual	ADJ
fcis-10369	3	26	connections	connection	NOUN
fcis-10369	3	27	with	with	ADP
fcis-10369	3	28	identity	identity	NOUN
fcis-10369	3	29	mappings	mapping	NOUN
fcis-10369	3	30	to	to	PART
fcis-10369	3	31	address	address	VERB
fcis-10369	3	32	the	the	DET
fcis-10369	3	33	problem	problem	NOUN
fcis-10369	3	34	of	of	ADP
fcis-10369	3	35	network	network	NOUN
fcis-10369	3	36	degradation	degradation	NOUN
fcis-10369	3	37	.	.	PUNCT
fcis-10369	4	1	the	the	DET
fcis-10369	4	2	multi	multi	ADJ
fcis-10369	4	3	-	-	ADJ
fcis-10369	4	4	scale	scale	ADJ
fcis-10369	4	5	convolution	convolution	NOUN
fcis-10369	4	6	captures	capture	VERB
fcis-10369	4	7	feature	feature	VERB
fcis-10369	4	8	representations	representation	NOUN
fcis-10369	4	9	at	at	ADP
fcis-10369	4	10	different	different	ADJ
fcis-10369	4	11	levels	level	NOUN
fcis-10369	4	12	,	,	PUNCT
fcis-10369	4	13	thereby	thereby	ADV
fcis-10369	4	14	improving	improve	VERB
fcis-10369	4	15	the	the	DET
fcis-10369	4	16	expressive	expressive	ADJ
fcis-10369	4	17	power	power	NOUN
fcis-10369	4	18	of	of	ADP
fcis-10369	4	19	the	the	DET
fcis-10369	4	20	model	model	NOUN
fcis-10369	4	21	.	.	PUNCT
fcis-10369	5	1	the	the	DET
fcis-10369	5	2	wgan	wgan	VERB
fcis-10369	5	3	-	-	PUNCT
fcis-10369	5	4	gp	gp	NOUN
fcis-10369	5	5	algorithm	algorithm	NOUN
fcis-10369	5	6	is	be	AUX
fcis-10369	5	7	employed	employ	VERB
fcis-10369	5	8	to	to	PART
fcis-10369	5	9	augment	augment	VERB
fcis-10369	5	10	the	the	DET
fcis-10369	5	11	minority	minority	NOUN
fcis-10369	5	12	samples	sample	NOUN
fcis-10369	5	13	and	and	CCONJ
fcis-10369	5	14	balance	balance	VERB
fcis-10369	5	15	the	the	DET
fcis-10369	5	16	dataset	dataset	NOUN
fcis-10369	5	17	.	.	PUNCT
fcis-10369	6	1	by	by	ADP
fcis-10369	6	2	performing	perform	VERB
fcis-10369	6	3	convolution	convolution	NOUN
fcis-10369	6	4	operations	operation	NOUN
fcis-10369	6	5	and	and	CCONJ
fcis-10369	6	6	extracting	extract	VERB
fcis-10369	6	7	local	local	ADJ
fcis-10369	6	8	window	window	NOUN
fcis-10369	6	9	features	feature	NOUN
fcis-10369	6	10	and	and	CCONJ
fcis-10369	6	11	global	global	ADJ
fcis-10369	6	12	features	feature	NOUN
fcis-10369	6	13	using	use	VERB
fcis-10369	6	14	bidirectional	bidirectional	ADJ
fcis-10369	6	15	lstm	lstm	ADJ
fcis-10369	6	16	units	unit	NOUN
fcis-10369	6	17	,	,	PUNCT
fcis-10369	6	18	the	the	DET
fcis-10369	6	19	model	model	NOUN
fcis-10369	6	20	effectively	effectively	ADV
fcis-10369	6	21	captures	capture	VERB
fcis-10369	6	22	temporal	temporal	ADJ
fcis-10369	6	23	information	information	NOUN
fcis-10369	6	24	and	and	CCONJ
fcis-10369	6	25	long	long	ADJ
fcis-10369	6	26	-	-	PUNCT
fcis-10369	6	27	term	term	NOUN
fcis-10369	6	28	dependencies	dependency	NOUN
fcis-10369	6	29	.	.	PUNCT
fcis-10369	7	1	experimental	experimental	ADJ
fcis-10369	7	2	results	result	NOUN
fcis-10369	7	3	demonstrate	demonstrate	VERB
fcis-10369	7	4	significant	significant	ADJ
fcis-10369	7	5	performance	performance	NOUN
fcis-10369	7	6	improvement	improvement	NOUN
fcis-10369	7	7	compared	compare	VERB
fcis-10369	7	8	to	to	ADP
fcis-10369	7	9	a	a	DET
fcis-10369	7	10	single	single	ADJ
fcis-10369	7	11	model	model	NOUN
fcis-10369	7	12	.	.	PUNCT
fcis-10369	8	1	the	the	DET
fcis-10369	8	2	msff	msff	PROPN
fcis-10369	8	3	model	model	NOUN
fcis-10369	8	4	achieves	achieve	VERB
fcis-10369	8	5	an	an	DET
fcis-10369	8	6	accuracy	accuracy	NOUN
fcis-10369	8	7	of	of	ADP
fcis-10369	8	8	99.50	99.50	NUM
fcis-10369	8	9	%	%	NOUN
fcis-10369	8	10	and	and	CCONJ
fcis-10369	8	11	94.73	94.73	NUM
fcis-10369	8	12	%	%	NOUN
fcis-10369	8	13	in	in	ADP
fcis-10369	8	14	binary	binary	ADJ
fcis-10369	8	15	classification	classification	NOUN
fcis-10369	8	16	experiments	experiment	NOUN
fcis-10369	8	17	on	on	ADP
fcis-10369	8	18	the	the	DET
fcis-10369	8	19	nsl	nsl	NOUN
fcis-10369	8	20	-	-	PUNCT
fcis-10369	8	21	kdd	kdd	PROPN
fcis-10369	8	22	and	and	CCONJ
fcis-10369	8	23	unsw	unsw	PROPN
fcis-10369	8	24	-	-	PUNCT
fcis-10369	8	25	nb15	nb15	PROPN
fcis-10369	8	26	datasets	dataset	NOUN
fcis-10369	8	27	,	,	PUNCT
fcis-10369	8	28	respectively	respectively	ADV
fcis-10369	8	29	,	,	PUNCT
fcis-10369	8	30	and	and	CCONJ
fcis-10369	8	31	an	an	DET
fcis-10369	8	32	accuracy	accuracy	NOUN
fcis-10369	8	33	of	of	ADP
fcis-10369	8	34	99.50	99.50	NUM
fcis-10369	8	35	%	%	NOUN
fcis-10369	8	36	and	and	CCONJ
fcis-10369	8	37	83.78	83.78	NUM
fcis-10369	8	38	%	%	NOUN
fcis-10369	8	39	in	in	ADP
fcis-10369	8	40	multi	multi	ADJ
fcis-10369	8	41	-	-	ADJ
fcis-10369	8	42	class	class	ADJ
fcis-10369	8	43	classification	classification	NOUN
fcis-10369	8	44	experiments	experiment	NOUN
fcis-10369	8	45	.	.	PUNCT
fcis-10369	9	1	keywords	keyword	NOUN
fcis-10369	9	2	:	:	PUNCT
fcis-10369	9	3	intrusion	intrusion	NOUN
fcis-10369	9	4	detection	detection	NOUN
fcis-10369	9	5	;	;	PUNCT
fcis-10369	9	6	multi	multi	ADJ
fcis-10369	9	7	-	-	ADJ
fcis-10369	9	8	scale	scale	ADJ
fcis-10369	9	9	1d	1d	NUM
fcis-10369	9	10	convolution	convolution	NOUN
fcis-10369	9	11	;	;	PUNCT
fcis-10369	9	12	bidirectional	bidirectional	ADJ
fcis-10369	9	13	long	long	ADJ
fcis-10369	9	14	short	short	ADJ
fcis-10369	9	15	-	-	PUNCT
fcis-10369	9	16	term	term	NOUN
fcis-10369	9	17	memory	memory	NOUN
fcis-10369	9	18	network	network	NOUN
fcis-10369	9	19	;	;	PUNCT
fcis-10369	9	20	residual	residual	ADJ
fcis-10369	9	21	connections	connection	NOUN
fcis-10369	9	22	.	.	PUNCT
fcis-10369	10	1	1	1	X
fcis-10369	10	2	.	.	X
fcis-10369	10	3	introduction	introduction	NOUN
fcis-10369	10	4	since	since	SCONJ
fcis-10369	10	5	the	the	DET
fcis-10369	10	6	popularization	popularization	NOUN
fcis-10369	10	7	of	of	ADP
fcis-10369	10	8	the	the	DET
fcis-10369	10	9	internet	internet	NOUN
fcis-10369	10	10	,	,	PUNCT
fcis-10369	10	11	network	network	NOUN
fcis-10369	10	12	security	security	NOUN
fcis-10369	10	13	has	have	AUX
fcis-10369	10	14	been	be	AUX
fcis-10369	10	15	a	a	DET
fcis-10369	10	16	hot	hot	ADJ
fcis-10369	10	17	topic	topic	NOUN
fcis-10369	10	18	of	of	ADP
fcis-10369	10	19	concern	concern	NOUN
fcis-10369	10	20	.	.	PUNCT
fcis-10369	11	1	at	at	ADP
fcis-10369	11	2	the	the	DET
fcis-10369	11	3	end	end	NOUN
fcis-10369	11	4	of	of	ADP
fcis-10369	11	5	2020	2020	NUM
fcis-10369	11	6	,	,	PUNCT
fcis-10369	11	7	hackers	hacker	NOUN
fcis-10369	11	8	exploited	exploit	VERB
fcis-10369	11	9	vulnerabilities	vulnerability	NOUN
fcis-10369	11	10	in	in	ADP
fcis-10369	11	11	the	the	DET
fcis-10369	11	12	supply	supply	NOUN
fcis-10369	11	13	chain	chain	NOUN
fcis-10369	11	14	of	of	ADP
fcis-10369	11	15	solarwinds	solarwind	NOUN
fcis-10369	11	16	,	,	PUNCT
fcis-10369	11	17	implanting	implant	VERB
fcis-10369	11	18	malicious	malicious	ADJ
fcis-10369	11	19	software	software	NOUN
fcis-10369	11	20	and	and	CCONJ
fcis-10369	11	21	resulting	result	VERB
fcis-10369	11	22	in	in	ADP
fcis-10369	11	23	large	large	ADJ
fcis-10369	11	24	-	-	PUNCT
fcis-10369	11	25	scale	scale	NOUN
fcis-10369	11	26	data	datum	NOUN
fcis-10369	11	27	breaches	breach	NOUN
fcis-10369	11	28	.	.	PUNCT
fcis-10369	12	1	in	in	ADP
fcis-10369	12	2	late	late	ADJ
fcis-10369	12	3	2021	2021	NUM
fcis-10369	12	4	,	,	PUNCT
fcis-10369	12	5	colonial	colonial	ADJ
fcis-10369	12	6	pipeline	pipeline	NOUN
fcis-10369	12	7	suffered	suffer	VERB
fcis-10369	12	8	a	a	DET
fcis-10369	12	9	cyberattack	cyberattack	NOUN
fcis-10369	12	10	that	that	PRON
fcis-10369	12	11	led	lead	VERB
fcis-10369	12	12	to	to	ADP
fcis-10369	12	13	the	the	DET
fcis-10369	12	14	shutdown	shutdown	NOUN
fcis-10369	12	15	of	of	ADP
fcis-10369	12	16	its	its	PRON
fcis-10369	12	17	pipeline	pipeline	NOUN
fcis-10369	12	18	system	system	NOUN
fcis-10369	12	19	for	for	ADP
fcis-10369	12	20	oil	oil	NOUN
fcis-10369	12	21	transportation	transportation	NOUN
fcis-10369	12	22	,	,	PUNCT
fcis-10369	12	23	causing	cause	VERB
fcis-10369	12	24	a	a	DET
fcis-10369	12	25	surge	surge	NOUN
fcis-10369	12	26	in	in	ADP
fcis-10369	12	27	oil	oil	NOUN
fcis-10369	12	28	prices	price	NOUN
fcis-10369	12	29	and	and	CCONJ
fcis-10369	12	30	public	public	ADJ
fcis-10369	12	31	panic	panic	NOUN
fcis-10369	12	32	in	in	ADP
fcis-10369	12	33	the	the	DET
fcis-10369	12	34	eastern	eastern	ADJ
fcis-10369	12	35	united	united	PROPN
fcis-10369	12	36	states	states	PROPN
fcis-10369	12	37	.	.	PUNCT
fcis-10369	13	1	also	also	ADV
fcis-10369	13	2	in	in	ADP
fcis-10369	13	3	the	the	DET
fcis-10369	13	4	same	same	ADJ
fcis-10369	13	5	year	year	NOUN
fcis-10369	13	6	,	,	PUNCT
fcis-10369	13	7	code	code	NOUN
fcis-10369	13	8	masters	master	NOUN
fcis-10369	13	9	,	,	PUNCT
fcis-10369	13	10	a	a	DET
fcis-10369	13	11	uk	uk	PROPN
fcis-10369	13	12	-	-	PUNCT
fcis-10369	13	13	based	base	VERB
fcis-10369	13	14	company	company	NOUN
fcis-10369	13	15	,	,	PUNCT
fcis-10369	13	16	fell	fall	VERB
fcis-10369	13	17	victim	victim	NOUN
fcis-10369	13	18	to	to	ADP
fcis-10369	13	19	a	a	DET
fcis-10369	13	20	ransomware	ransomware	NOUN
fcis-10369	13	21	attack	attack	NOUN
fcis-10369	13	22	,	,	PUNCT
fcis-10369	13	23	resulting	result	VERB
fcis-10369	13	24	in	in	ADP
fcis-10369	13	25	a	a	DET
fcis-10369	13	26	significant	significant	ADJ
fcis-10369	13	27	leak	leak	NOUN
fcis-10369	13	28	of	of	ADP
fcis-10369	13	29	user	user	NOUN
fcis-10369	13	30	's	's	PART
fcis-10369	13	31	private	private	ADJ
fcis-10369	13	32	data	datum	NOUN
fcis-10369	13	33	.	.	PUNCT
fcis-10369	14	1	therefore	therefore	ADV
fcis-10369	14	2	,	,	PUNCT
fcis-10369	14	3	in	in	ADP
fcis-10369	14	4	the	the	DET
fcis-10369	14	5	current	current	ADJ
fcis-10369	14	6	environment	environment	NOUN
fcis-10369	14	7	,	,	PUNCT
fcis-10369	14	8	it	it	PRON
fcis-10369	14	9	is	be	AUX
fcis-10369	14	10	an	an	DET
fcis-10369	14	11	urgent	urgent	ADJ
fcis-10369	14	12	problem	problem	NOUN
fcis-10369	14	13	to	to	PART
fcis-10369	14	14	effectively	effectively	ADV
fcis-10369	14	15	prevent	prevent	VERB
fcis-10369	14	16	and	and	CCONJ
fcis-10369	14	17	detect	detect	VERB
fcis-10369	14	18	network	network	NOUN
fcis-10369	14	19	attacks	attack	NOUN
fcis-10369	14	20	.	.	PUNCT
fcis-10369	15	1	intrusion	intrusion	NOUN
fcis-10369	15	2	detection	detection	NOUN
fcis-10369	15	3	systems	system	NOUN
fcis-10369	15	4	(	(	PUNCT
fcis-10369	15	5	ids	id	NOUN
fcis-10369	15	6	)	)	PUNCT
fcis-10369	15	7	monitor	monitor	NOUN
fcis-10369	15	8	network	network	NOUN
fcis-10369	15	9	data	datum	NOUN
fcis-10369	15	10	traffic	traffic	NOUN
fcis-10369	15	11	to	to	PART
fcis-10369	15	12	detect	detect	VERB
fcis-10369	15	13	abnormal	abnormal	ADJ
fcis-10369	15	14	flows	flow	NOUN
fcis-10369	15	15	and	and	CCONJ
fcis-10369	15	16	prevent	prevent	VERB
fcis-10369	15	17	network	network	NOUN
fcis-10369	15	18	intrusion	intrusion	NOUN
fcis-10369	15	19	activities	activity	NOUN
fcis-10369	15	20	.	.	PUNCT
fcis-10369	16	1	several	several	ADJ
fcis-10369	16	2	traditional	traditional	ADJ
fcis-10369	16	3	statistical	statistical	ADJ
fcis-10369	16	4	machine	machine	NOUN
fcis-10369	16	5	learning	learning	NOUN
fcis-10369	16	6	algorithms	algorithm	NOUN
fcis-10369	16	7	have	have	AUX
fcis-10369	16	8	been	be	AUX
fcis-10369	16	9	applied	apply	VERB
fcis-10369	16	10	to	to	ADP
fcis-10369	16	11	network	network	NOUN
fcis-10369	16	12	intrusion	intrusion	NOUN
fcis-10369	16	13	detection	detection	NOUN
fcis-10369	16	14	,	,	PUNCT
fcis-10369	16	15	such	such	ADJ
fcis-10369	16	16	as	as	ADP
fcis-10369	16	17	k	k	NOUN
fcis-10369	16	18	-	-	PUNCT
fcis-10369	16	19	nearest	near	ADJ
fcis-10369	16	20	neighbors	neighbor	NOUN
fcis-10369	16	21	[	[	X
fcis-10369	16	22	1	1	NUM
fcis-10369	16	23	-	-	SYM
fcis-10369	16	24	2	2	NUM
fcis-10369	16	25	]	]	PUNCT
fcis-10369	16	26	,	,	PUNCT
fcis-10369	16	27	support	support	VERB
fcis-10369	16	28	vector	vector	NOUN
fcis-10369	16	29	machines	machine	NOUN
fcis-10369	17	1	[	[	X
fcis-10369	17	2	3	3	NUM
fcis-10369	17	3	-	-	SYM
fcis-10369	17	4	4	4	NUM
fcis-10369	17	5	]	]	PUNCT
fcis-10369	17	6	,	,	PUNCT
fcis-10369	17	7	naive	naive	ADJ
fcis-10369	17	8	bayes	bayes	NOUN
fcis-10369	18	1	[	[	X
fcis-10369	18	2	5	5	NUM
fcis-10369	18	3	]	]	PUNCT
fcis-10369	18	4	,	,	PUNCT
fcis-10369	18	5	random	random	ADJ
fcis-10369	18	6	forests	forest	NOUN
fcis-10369	18	7	[	[	X
fcis-10369	18	8	6	6	NUM
fcis-10369	18	9	-	-	SYM
fcis-10369	18	10	7	7	NUM
fcis-10369	18	11	]	]	PUNCT
fcis-10369	18	12	,	,	PUNCT
fcis-10369	18	13	etc	etc	X
fcis-10369	18	14	.	.	X
fcis-10369	19	1	these	these	DET
fcis-10369	19	2	traditional	traditional	ADJ
fcis-10369	19	3	machine	machine	NOUN
fcis-10369	19	4	learning	learning	NOUN
fcis-10369	19	5	approaches	approach	NOUN
fcis-10369	19	6	often	often	ADV
fcis-10369	19	7	focus	focus	VERB
fcis-10369	19	8	on	on	ADP
fcis-10369	19	9	the	the	DET
fcis-10369	19	10	relationship	relationship	NOUN
fcis-10369	19	11	between	between	ADP
fcis-10369	19	12	individual	individual	ADJ
fcis-10369	19	13	features	feature	NOUN
fcis-10369	19	14	and	and	CCONJ
fcis-10369	19	15	the	the	DET
fcis-10369	19	16	outcome	outcome	NOUN
fcis-10369	19	17	during	during	ADP
fcis-10369	19	18	model	model	NOUN
fcis-10369	19	19	training	training	NOUN
fcis-10369	19	20	,	,	PUNCT
fcis-10369	19	21	while	while	SCONJ
fcis-10369	19	22	neglecting	neglect	VERB
fcis-10369	19	23	the	the	DET
fcis-10369	19	24	relationships	relationship	NOUN
fcis-10369	19	25	between	between	ADP
fcis-10369	19	26	different	different	ADJ
fcis-10369	19	27	features	feature	NOUN
fcis-10369	19	28	and	and	CCONJ
fcis-10369	19	29	often	often	ADV
fcis-10369	19	30	requiring	require	VERB
fcis-10369	19	31	manual	manual	ADJ
fcis-10369	19	32	feature	feature	NOUN
fcis-10369	19	33	selection	selection	NOUN
fcis-10369	19	34	.	.	PUNCT
fcis-10369	20	1	in	in	ADP
fcis-10369	20	2	recent	recent	ADJ
fcis-10369	20	3	years	year	NOUN
fcis-10369	20	4	,	,	PUNCT
fcis-10369	20	5	the	the	DET
fcis-10369	20	6	diversity	diversity	NOUN
fcis-10369	20	7	and	and	CCONJ
fcis-10369	20	8	complexity	complexity	NOUN
fcis-10369	20	9	of	of	ADP
fcis-10369	20	10	network	network	NOUN
fcis-10369	20	11	traffic	traffic	NOUN
fcis-10369	20	12	data	datum	NOUN
fcis-10369	20	13	have	have	AUX
fcis-10369	20	14	grown	grow	VERB
fcis-10369	20	15	exponentially	exponentially	ADV
fcis-10369	20	16	,	,	PUNCT
fcis-10369	20	17	rendering	render	VERB
fcis-10369	20	18	traditional	traditional	ADJ
fcis-10369	20	19	statistical	statistical	ADJ
fcis-10369	20	20	machine	machine	NOUN
fcis-10369	20	21	learning	learning	NOUN
fcis-10369	20	22	models	model	NOUN
fcis-10369	20	23	inadequate	inadequate	ADJ
fcis-10369	20	24	for	for	ADP
fcis-10369	20	25	the	the	DET
fcis-10369	20	26	current	current	ADJ
fcis-10369	20	27	data	datum	NOUN
fcis-10369	20	28	volume	volume	NOUN
fcis-10369	20	29	.	.	PUNCT
fcis-10369	21	1	deep	deep	ADJ
fcis-10369	21	2	learning	learning	NOUN
fcis-10369	21	3	methods	method	NOUN
fcis-10369	21	4	,	,	PUNCT
fcis-10369	21	5	with	with	ADP
fcis-10369	21	6	their	their	PRON
fcis-10369	21	7	superiority	superiority	NOUN
fcis-10369	21	8	in	in	ADP
fcis-10369	21	9	handling	handle	VERB
fcis-10369	21	10	large	large	ADJ
fcis-10369	21	11	amounts	amount	NOUN
fcis-10369	21	12	of	of	ADP
fcis-10369	21	13	nonlinear	nonlinear	ADJ
fcis-10369	21	14	data	datum	NOUN
fcis-10369	21	15	and	and	CCONJ
fcis-10369	21	16	extracting	extract	VERB
fcis-10369	21	17	advanced	advanced	ADJ
fcis-10369	21	18	features	feature	NOUN
fcis-10369	21	19	from	from	ADP
fcis-10369	21	20	raw	raw	ADJ
fcis-10369	21	21	data	datum	NOUN
fcis-10369	21	22	,	,	PUNCT
fcis-10369	21	23	have	have	AUX
fcis-10369	21	24	been	be	AUX
fcis-10369	21	25	widely	widely	ADV
fcis-10369	21	26	adopted	adopt	VERB
fcis-10369	21	27	in	in	ADP
fcis-10369	21	28	network	network	NOUN
fcis-10369	21	29	intrusion	intrusion	NOUN
fcis-10369	21	30	detection	detection	NOUN
fcis-10369	21	31	.	.	PUNCT
fcis-10369	22	1	after	after	ADP
fcis-10369	22	2	being	be	AUX
fcis-10369	22	3	processed	process	VERB
fcis-10369	22	4	by	by	ADP
fcis-10369	22	5	deep	deep	ADJ
fcis-10369	22	6	learning	learning	NOUN
fcis-10369	22	7	methods	method	NOUN
fcis-10369	22	8	,	,	PUNCT
fcis-10369	22	9	network	network	NOUN
fcis-10369	22	10	traffic	traffic	NOUN
fcis-10369	22	11	data	datum	NOUN
fcis-10369	22	12	can	can	AUX
fcis-10369	22	13	yield	yield	VERB
fcis-10369	22	14	advanced	advanced	ADJ
fcis-10369	22	15	features	feature	NOUN
fcis-10369	22	16	that	that	PRON
fcis-10369	22	17	are	be	AUX
fcis-10369	22	18	potentially	potentially	ADV
fcis-10369	22	19	related	relate	VERB
fcis-10369	22	20	to	to	ADP
fcis-10369	22	21	attack	attack	NOUN
fcis-10369	22	22	behaviors	behavior	NOUN
fcis-10369	22	23	,	,	PUNCT
fcis-10369	22	24	such	such	ADJ
fcis-10369	22	25	as	as	ADP
fcis-10369	22	26	malicious	malicious	ADJ
fcis-10369	22	27	software	software	NOUN
fcis-10369	22	28	protocols	protocol	NOUN
fcis-10369	22	29	,	,	PUNCT
fcis-10369	22	30	ports	port	NOUN
fcis-10369	22	31	,	,	PUNCT
fcis-10369	22	32	packet	packet	NOUN
fcis-10369	22	33	sizes	size	NOUN
fcis-10369	22	34	,	,	PUNCT
fcis-10369	22	35	etc	etc	X
fcis-10369	22	36	.	.	X
fcis-10369	23	1	2	2	X
fcis-10369	23	2	.	.	NUM
fcis-10369	23	3	related	relate	VERB
fcis-10369	23	4	work	work	NOUN
fcis-10369	23	5	compared	compare	VERB
fcis-10369	23	6	to	to	ADP
fcis-10369	23	7	traditional	traditional	ADJ
fcis-10369	23	8	feature	feature	NOUN
fcis-10369	23	9	engineering	engineering	NOUN
fcis-10369	23	10	methods	method	NOUN
fcis-10369	23	11	,	,	PUNCT
fcis-10369	23	12	deep	deep	ADJ
fcis-10369	23	13	learning	learning	NOUN
fcis-10369	23	14	methods	method	NOUN
fcis-10369	23	15	can	can	AUX
fcis-10369	23	16	automatically	automatically	ADV
fcis-10369	23	17	learn	learn	VERB
fcis-10369	23	18	more	more	ADJ
fcis-10369	23	19	discriminative	discriminative	NOUN
fcis-10369	23	20	features	feature	NOUN
fcis-10369	23	21	from	from	ADP
fcis-10369	23	22	raw	raw	ADJ
fcis-10369	23	23	data	datum	NOUN
fcis-10369	23	24	,	,	PUNCT
fcis-10369	23	25	thereby	thereby	ADV
fcis-10369	23	26	improving	improve	VERB
fcis-10369	23	27	the	the	DET
fcis-10369	23	28	accuracy	accuracy	NOUN
fcis-10369	23	29	and	and	CCONJ
fcis-10369	23	30	efficiency	efficiency	NOUN
fcis-10369	23	31	of	of	ADP
fcis-10369	23	32	intrusion	intrusion	NOUN
fcis-10369	23	33	detection	detection	NOUN
fcis-10369	23	34	.	.	PUNCT
fcis-10369	24	1	shu	shu	PROPN
fcis-10369	24	2	hao	hao	PROPN
fcis-10369	25	1	[	[	X
fcis-10369	25	2	8	8	NUM
fcis-10369	25	3	]	]	PUNCT
fcis-10369	25	4	proposed	propose	VERB
fcis-10369	25	5	an	an	DET
fcis-10369	25	6	intrusion	intrusion	NOUN
fcis-10369	25	7	detection	detection	NOUN
fcis-10369	25	8	method	method	NOUN
fcis-10369	25	9	that	that	PRON
fcis-10369	25	10	combines	combine	VERB
fcis-10369	25	11	bidirectional	bidirectional	ADJ
fcis-10369	25	12	long	long	ADJ
fcis-10369	25	13	short	short	ADJ
fcis-10369	25	14	-	-	PUNCT
fcis-10369	25	15	term	term	NOUN
fcis-10369	25	16	memory	memory	NOUN
fcis-10369	25	17	(	(	PUNCT
fcis-10369	25	18	bilstm	bilstm	NOUN
fcis-10369	25	19	)	)	PUNCT
fcis-10369	25	20	and	and	CCONJ
fcis-10369	25	21	attention	attention	NOUN
fcis-10369	25	22	mechanism	mechanism	NOUN
fcis-10369	25	23	to	to	PART
fcis-10369	25	24	improve	improve	VERB
fcis-10369	25	25	the	the	DET
fcis-10369	25	26	detection	detection	NOUN
fcis-10369	25	27	performance	performance	NOUN
fcis-10369	25	28	on	on	ADP
fcis-10369	25	29	nsl	nsl	PROPN
fcis-10369	25	30	-	-	PUNCT
fcis-10369	25	31	kdd	kdd	PROPN
fcis-10369	25	32	dataset	dataset	NOUN
fcis-10369	25	33	.	.	PUNCT
fcis-10369	26	1	compared	compare	VERB
fcis-10369	26	2	to	to	ADP
fcis-10369	26	3	the	the	DET
fcis-10369	26	4	single	single	ADJ
fcis-10369	26	5	bilstm	bilstm	NOUN
fcis-10369	26	6	method	method	NOUN
fcis-10369	26	7	,	,	PUNCT
fcis-10369	26	8	it	it	PRON
fcis-10369	26	9	achieved	achieve	VERB
fcis-10369	26	10	improvements	improvement	NOUN
fcis-10369	26	11	in	in	ADP
fcis-10369	26	12	accuracy	accuracy	NOUN
fcis-10369	26	13	and	and	CCONJ
fcis-10369	26	14	f1	f1	NOUN
fcis-10369	26	15	score	score	NOUN
fcis-10369	26	16	,	,	PUNCT
fcis-10369	26	17	but	but	CCONJ
fcis-10369	26	18	the	the	DET
fcis-10369	26	19	multi	multi	ADJ
fcis-10369	26	20	-	-	ADJ
fcis-10369	26	21	class	class	ADJ
fcis-10369	26	22	classification	classification	NOUN
fcis-10369	26	23	results	result	NOUN
fcis-10369	26	24	were	be	AUX
fcis-10369	26	25	not	not	PART
fcis-10369	26	26	ideal	ideal	ADJ
fcis-10369	26	27	.	.	PUNCT
fcis-10369	27	1	li	li	PROPN
fcis-10369	27	2	haitao	haitao	PROPN
fcis-10369	27	3	et	et	PROPN
fcis-10369	27	4	al	al	PROPN
fcis-10369	27	5	.	.	PUNCT
fcis-10369	28	1	[	[	X
fcis-10369	28	2	9	9	NUM
fcis-10369	28	3	]	]	PUNCT
fcis-10369	28	4	proposed	propose	VERB
fcis-10369	28	5	a	a	DET
fcis-10369	28	6	semi	semi	ADJ
fcis-10369	28	7	-	-	ADJ
fcis-10369	28	8	gru	gru	VERB
fcis-10369	28	9	-	-	PUNCT
fcis-10369	28	10	based	base	VERB
fcis-10369	28	11	traffic	traffic	NOUN
fcis-10369	28	12	anomaly	anomaly	NOUN
fcis-10369	28	13	detection	detection	NOUN
fcis-10369	28	14	method	method	NOUN
fcis-10369	28	15	that	that	PRON
fcis-10369	28	16	combines	combine	VERB
fcis-10369	28	17	multiple	multiple	ADJ
fcis-10369	28	18	layers	layer	NOUN
fcis-10369	28	19	of	of	ADP
fcis-10369	28	20	bidirectional	bidirectional	ADJ
fcis-10369	28	21	gated	gate	VERB
fcis-10369	28	22	recurrent	recurrent	ADJ
fcis-10369	28	23	units	unit	NOUN
fcis-10369	28	24	(	(	PUNCT
fcis-10369	28	25	gru	gru	NOUN
fcis-10369	28	26	)	)	PUNCT
fcis-10369	28	27	and	and	CCONJ
fcis-10369	28	28	an	an	DET
fcis-10369	28	29	improved	improved	ADJ
fcis-10369	28	30	feedforward	feedforward	NOUN
fcis-10369	28	31	network	network	NOUN
fcis-10369	28	32	,	,	PUNCT
fcis-10369	28	33	effectively	effectively	ADV
fcis-10369	28	34	improving	improve	VERB
fcis-10369	28	35	the	the	DET
fcis-10369	28	36	f1	f1	ADJ
fcis-10369	28	37	score	score	NOUN
fcis-10369	28	38	but	but	CCONJ
fcis-10369	28	39	with	with	ADP
fcis-10369	28	40	lower	low	ADJ
fcis-10369	28	41	accuracy	accuracy	NOUN
fcis-10369	28	42	.	.	PUNCT
fcis-10369	29	1	muhuri	muhuri	PROPN
fcis-10369	29	2	et	et	PROPN
fcis-10369	29	3	al	al	PROPN
fcis-10369	29	4	.	.	PUNCT
fcis-10369	30	1	[	[	X
fcis-10369	30	2	10	10	NUM
fcis-10369	30	3	]	]	PUNCT
fcis-10369	30	4	proposed	propose	VERB
fcis-10369	30	5	a	a	DET
fcis-10369	30	6	combination	combination	NOUN
fcis-10369	30	7	of	of	ADP
fcis-10369	30	8	long	long	ADJ
fcis-10369	30	9	short	short	ADJ
fcis-10369	30	10	-	-	PUNCT
fcis-10369	30	11	term	term	NOUN
fcis-10369	30	12	memory	memory	NOUN
fcis-10369	30	13	(	(	PUNCT
fcis-10369	30	14	lstm	lstm	NOUN
fcis-10369	30	15	)	)	PUNCT
fcis-10369	30	16	and	and	CCONJ
fcis-10369	30	17	genetic	genetic	ADJ
fcis-10369	30	18	algorithm	algorithm	NOUN
fcis-10369	30	19	(	(	PUNCT
fcis-10369	30	20	ga	ga	NOUN
fcis-10369	30	21	)	)	PUNCT
fcis-10369	30	22	for	for	ADP
fcis-10369	30	23	feature	feature	NOUN
fcis-10369	30	24	selection	selection	NOUN
fcis-10369	30	25	,	,	PUNCT
fcis-10369	30	26	achieving	achieve	VERB
fcis-10369	30	27	an	an	DET
fcis-10369	30	28	accuracy	accuracy	NOUN
fcis-10369	30	29	of	of	ADP
fcis-10369	30	30	93.88	93.88	NUM
fcis-10369	30	31	%	%	NOUN
fcis-10369	30	32	on	on	ADP
fcis-10369	30	33	nsl	nsl	PROPN
fcis-10369	30	34	-	-	PUNCT
fcis-10369	30	35	kdd	kdd	NOUN
fcis-10369	30	36	,	,	PUNCT
fcis-10369	30	37	but	but	CCONJ
fcis-10369	30	38	they	they	PRON
fcis-10369	30	39	did	do	AUX
fcis-10369	30	40	not	not	PART
fcis-10369	30	41	provide	provide	VERB
fcis-10369	30	42	the	the	DET
fcis-10369	30	43	detection	detection	NOUN
fcis-10369	30	44	rate	rate	NOUN
fcis-10369	30	45	.	.	PUNCT
fcis-10369	30	46	yu	yu	PROPN
fcis-10369	31	1	[	[	X
fcis-10369	31	2	11	11	NUM
fcis-10369	31	3	]	]	PUNCT
fcis-10369	31	4	combined	combine	VERB
fcis-10369	31	5	deep	deep	ADJ
fcis-10369	31	6	belief	belief	NOUN
fcis-10369	31	7	network	network	NOUN
fcis-10369	31	8	(	(	PUNCT
fcis-10369	31	9	dbn	dbn	PROPN
fcis-10369	31	10	)	)	PUNCT
fcis-10369	31	11	and	and	CCONJ
fcis-10369	31	12	bidirectional	bidirectional	ADJ
fcis-10369	31	13	gated	gate	VERB
fcis-10369	31	14	recurrent	recurrent	ADJ
fcis-10369	31	15	units	unit	NOUN
fcis-10369	31	16	(	(	PUNCT
fcis-10369	31	17	bigru	bigru	NOUN
fcis-10369	31	18	)	)	PUNCT
fcis-10369	31	19	in	in	ADP
fcis-10369	31	20	an	an	DET
fcis-10369	31	21	algorithm	algorithm	NOUN
fcis-10369	31	22	model	model	NOUN
fcis-10369	31	23	(	(	PUNCT
fcis-10369	31	24	dbn	dbn	NOUN
fcis-10369	31	25	-	-	PUNCT
fcis-10369	31	26	bigru	bigru	NOUN
fcis-10369	31	27	)	)	PUNCT
fcis-10369	31	28	and	and	CCONJ
fcis-10369	31	29	processed	process	VERB
fcis-10369	31	30	the	the	DET
fcis-10369	31	31	raw	raw	ADJ
fcis-10369	31	32	feature	feature	NOUN
fcis-10369	31	33	analysis	analysis	NOUN
fcis-10369	31	34	files	file	NOUN
fcis-10369	31	35	of	of	ADP
fcis-10369	31	36	cicids2017	cicids2017	NOUN
fcis-10369	31	37	into	into	ADP
fcis-10369	31	38	well	well	ADV
fcis-10369	31	39	-	-	PUNCT
fcis-10369	31	40	structured	structure	VERB
fcis-10369	31	41	temporal	temporal	ADJ
fcis-10369	31	42	data	datum	NOUN
fcis-10369	31	43	,	,	PUNCT
fcis-10369	31	44	but	but	CCONJ
fcis-10369	31	45	no	no	DET
fcis-10369	31	46	multiclass	multiclass	ADJ
fcis-10369	31	47	experiments	experiment	NOUN
fcis-10369	31	48	were	be	AUX
fcis-10369	31	49	conducted	conduct	VERB
fcis-10369	31	50	.	.	PUNCT
fcis-10369	32	1	khan	khan	PROPN
fcis-10369	33	1	[	[	X
fcis-10369	33	2	12	12	NUM
fcis-10369	33	3	]	]	PUNCT
fcis-10369	33	4	proposed	propose	VERB
fcis-10369	33	5	a	a	DET
fcis-10369	33	6	hybrid	hybrid	ADJ
fcis-10369	33	7	convolutional	convolutional	ADJ
fcis-10369	33	8	recurrent	recurrent	ADJ
fcis-10369	33	9	neural	neural	ADJ
fcis-10369	33	10	network	network	NOUN
fcis-10369	33	11	intrusion	intrusion	NOUN
fcis-10369	33	12	detection	detection	NOUN
fcis-10369	33	13	system	system	NOUN
fcis-10369	33	14	(	(	PUNCT
fcis-10369	33	15	hcrnnids	hcrnnid	NOUN
fcis-10369	33	16	)	)	PUNCT
fcis-10369	33	17	where	where	SCONJ
fcis-10369	33	18	convolutional	convolutional	ADJ
fcis-10369	33	19	neural	neural	ADJ
fcis-10369	33	20	network	network	NOUN
fcis-10369	33	21	(	(	PUNCT
fcis-10369	33	22	cnn	cnn	PROPN
fcis-10369	33	23	)	)	PUNCT
fcis-10369	33	24	captures	capture	VERB
fcis-10369	33	25	local	local	ADJ
fcis-10369	33	26	features	feature	NOUN
fcis-10369	33	27	through	through	ADP
fcis-10369	33	28	convolution	convolution	NOUN
fcis-10369	33	29	and	and	CCONJ
fcis-10369	33	30	recurrent	recurrent	ADJ
fcis-10369	33	31	neural	neural	ADJ
fcis-10369	33	32	network	network	NOUN
fcis-10369	33	33	(	(	PUNCT
fcis-10369	33	34	rnn	rnn	PROPN
fcis-10369	33	35	)	)	PUNCT
fcis-10369	33	36	captures	capture	VERB
fcis-10369	33	37	temporal	temporal	ADJ
fcis-10369	33	38	features	feature	NOUN
fcis-10369	33	39	to	to	PART
fcis-10369	33	40	improve	improve	VERB
fcis-10369	33	41	the	the	DET
fcis-10369	33	42	performance	performance	NOUN
fcis-10369	33	43	and	and	CCONJ
fcis-10369	33	44	prediction	prediction	NOUN
fcis-10369	33	45	of	of	ADP
fcis-10369	33	46	the	the	DET
fcis-10369	33	47	intrusion	intrusion	NOUN
fcis-10369	33	48	detection	detection	NOUN
fcis-10369	33	49	system	system	NOUN
fcis-10369	33	50	,	,	PUNCT
fcis-10369	33	51	achieving	achieve	VERB
fcis-10369	33	52	improvements	improvement	NOUN
fcis-10369	33	53	in	in	ADP
fcis-10369	33	54	precision	precision	NOUN
fcis-10369	33	55	and	and	CCONJ
fcis-10369	33	56	other	other	ADJ
fcis-10369	33	57	metrics	metric	NOUN
fcis-10369	33	58	,	,	PUNCT
fcis-10369	33	59	but	but	CCONJ
fcis-10369	33	60	again	again	ADV
fcis-10369	33	61	,	,	PUNCT
fcis-10369	33	62	no	no	DET
fcis-10369	33	63	multi	multi	ADJ
fcis-10369	33	64	-	-	ADJ
fcis-10369	33	65	class	class	ADJ
fcis-10369	33	66	experiments	experiment	NOUN
fcis-10369	33	67	were	be	AUX
fcis-10369	33	68	conducted	conduct	VERB
fcis-10369	33	69	.	.	PUNCT
fcis-10369	34	1	the	the	DET
fcis-10369	34	2	majority	majority	NOUN
fcis-10369	34	3	of	of	ADP
fcis-10369	34	4	machine	machine	NOUN
fcis-10369	34	5	learning	learning	NOUN
fcis-10369	34	6	-	-	PUNCT
fcis-10369	34	7	based	base	VERB
fcis-10369	34	8	algorithms	algorithm	NOUN
fcis-10369	34	9	are	be	AUX
fcis-10369	34	10	built	build	VERB
fcis-10369	34	11	on	on	ADP
fcis-10369	34	12	the	the	DET
fcis-10369	34	13	assumption	assumption	NOUN
fcis-10369	34	14	of	of	ADP
fcis-10369	34	15	sufficient	sufficient	ADJ
fcis-10369	34	16	learning	learning	NOUN
fcis-10369	34	17	from	from	ADP
fcis-10369	34	18	a	a	DET
fcis-10369	34	19	large	large	ADJ
fcis-10369	34	20	amount	amount	NOUN
fcis-10369	34	21	of	of	ADP
fcis-10369	34	22	raw	raw	ADJ
fcis-10369	34	23	data	datum	NOUN
fcis-10369	34	24	.	.	PUNCT
fcis-10369	35	1	however	however	ADV
fcis-10369	35	2	,	,	PUNCT
fcis-10369	35	3	the	the	DET
fcis-10369	35	4	problem	problem	NOUN
fcis-10369	35	5	of	of	ADP
fcis-10369	35	6	imbalanced	imbalanced	ADJ
fcis-10369	35	7	125	125	NUM
fcis-10369	35	8	data	datum	NOUN
fcis-10369	35	9	distribution	distribution	NOUN
fcis-10369	35	10	is	be	AUX
fcis-10369	35	11	prevalent	prevalent	ADJ
fcis-10369	35	12	[	[	X
fcis-10369	35	13	13	13	NUM
fcis-10369	35	14	]	]	PUNCT
fcis-10369	35	15	.	.	PUNCT
fcis-10369	36	1	insufficient	insufficient	ADJ
fcis-10369	36	2	feature	feature	NOUN
fcis-10369	36	3	learning	learn	VERB
fcis-10369	36	4	from	from	ADP
fcis-10369	36	5	the	the	DET
fcis-10369	36	6	minority	minority	NOUN
fcis-10369	36	7	class	class	NOUN
fcis-10369	36	8	samples	sample	NOUN
fcis-10369	36	9	leads	lead	VERB
fcis-10369	36	10	to	to	ADP
fcis-10369	36	11	lower	low	ADJ
fcis-10369	36	12	detection	detection	NOUN
fcis-10369	36	13	rates	rate	NOUN
fcis-10369	36	14	for	for	ADP
fcis-10369	36	15	these	these	DET
fcis-10369	36	16	classes	class	NOUN
fcis-10369	36	17	,	,	PUNCT
fcis-10369	36	18	which	which	PRON
fcis-10369	36	19	affects	affect	VERB
fcis-10369	36	20	the	the	DET
fcis-10369	36	21	performance	performance	NOUN
fcis-10369	36	22	of	of	ADP
fcis-10369	36	23	intrusion	intrusion	NOUN
fcis-10369	36	24	detection	detection	NOUN
fcis-10369	36	25	models	model	NOUN
fcis-10369	36	26	and	and	CCONJ
fcis-10369	36	27	may	may	AUX
fcis-10369	36	28	result	result	VERB
fcis-10369	36	29	in	in	ADP
fcis-10369	36	30	highly	highly	ADV
fcis-10369	36	31	threatening	threatening	ADJ
fcis-10369	36	32	attack	attack	NOUN
fcis-10369	36	33	behaviors	behavior	NOUN
fcis-10369	36	34	being	be	AUX
fcis-10369	36	35	misclassified	misclassifie	VERB
fcis-10369	36	36	as	as	ADP
fcis-10369	36	37	normal	normal	ADJ
fcis-10369	36	38	behavior	behavior	NOUN
fcis-10369	36	39	,	,	PUNCT
fcis-10369	36	40	thereby	thereby	ADV
fcis-10369	36	41	adversely	adversely	ADV
fcis-10369	36	42	affecting	affect	VERB
fcis-10369	36	43	the	the	DET
fcis-10369	36	44	devices	device	NOUN
fcis-10369	36	45	.	.	PUNCT
fcis-10369	37	1	zhang	zhang	PROPN
fcis-10369	38	1	[	[	X
fcis-10369	38	2	14	14	NUM
fcis-10369	38	3	]	]	PUNCT
fcis-10369	38	4	studied	study	VERB
fcis-10369	38	5	the	the	DET
fcis-10369	38	6	combination	combination	NOUN
fcis-10369	38	7	of	of	ADP
fcis-10369	38	8	gaussian	gaussian	ADJ
fcis-10369	38	9	mixture	mixture	NOUN
fcis-10369	38	10	model	model	NOUN
fcis-10369	38	11	-	-	PUNCT
fcis-10369	38	12	based	base	VERB
fcis-10369	38	13	smote	smote	NOUN
fcis-10369	38	14	oversampling	oversample	VERB
fcis-10369	38	15	and	and	CCONJ
fcis-10369	38	16	clustering	clustering	NOUN
fcis-10369	38	17	-	-	PUNCT
fcis-10369	38	18	based	base	VERB
fcis-10369	38	19	undersampling	undersample	VERB
fcis-10369	38	20	on	on	ADP
fcis-10369	38	21	the	the	DET
fcis-10369	38	22	cicids2017	cicids2017	NOUN
fcis-10369	38	23	dataset	dataset	NOUN
fcis-10369	38	24	.	.	PUNCT
fcis-10369	39	1	although	although	SCONJ
fcis-10369	39	2	their	their	PRON
fcis-10369	39	3	model	model	NOUN
fcis-10369	39	4	achieved	achieve	VERB
fcis-10369	39	5	good	good	ADJ
fcis-10369	39	6	performance	performance	NOUN
fcis-10369	39	7	,	,	PUNCT
fcis-10369	39	8	they	they	PRON
fcis-10369	39	9	did	do	AUX
fcis-10369	39	10	not	not	PART
fcis-10369	39	11	provide	provide	VERB
fcis-10369	39	12	the	the	DET
fcis-10369	39	13	f1	f1	NOUN
fcis-10369	39	14	score	score	NOUN
fcis-10369	39	15	.	.	PUNCT
fcis-10369	40	1	jiang	jiang	PROPN
fcis-10369	41	1	[	[	X
fcis-10369	41	2	15	15	NUM
fcis-10369	41	3	]	]	PUNCT
fcis-10369	41	4	combined	combine	VERB
fcis-10369	41	5	one	one	NUM
fcis-10369	41	6	-	-	PUNCT
fcis-10369	41	7	sided	sided	ADJ
fcis-10369	41	8	selection	selection	NOUN
fcis-10369	41	9	(	(	PUNCT
fcis-10369	41	10	oss	oss	NOUN
fcis-10369	41	11	)	)	PUNCT
fcis-10369	41	12	with	with	ADP
fcis-10369	41	13	synthetic	synthetic	ADJ
fcis-10369	41	14	minority	minority	NOUN
fcis-10369	41	15	oversampling	oversample	VERB
fcis-10369	41	16	technique	technique	NOUN
fcis-10369	41	17	(	(	PUNCT
fcis-10369	41	18	smote	smote	NOUN
fcis-10369	41	19	)	)	PUNCT
fcis-10369	41	20	and	and	CCONJ
fcis-10369	41	21	incorporated	incorporate	VERB
fcis-10369	41	22	a	a	DET
fcis-10369	41	23	deep	deep	ADJ
fcis-10369	41	24	hierarchical	hierarchical	ADJ
fcis-10369	41	25	model	model	NOUN
fcis-10369	41	26	for	for	ADP
fcis-10369	41	27	spatialtemporal	spatialtemporal	ADJ
fcis-10369	41	28	feature	feature	NOUN
fcis-10369	41	29	extraction	extraction	NOUN
fcis-10369	41	30	.	.	PUNCT
fcis-10369	42	1	this	this	DET
fcis-10369	42	2	method	method	NOUN
fcis-10369	42	3	improved	improve	VERB
fcis-10369	42	4	the	the	DET
fcis-10369	42	5	metrics	metric	NOUN
fcis-10369	42	6	for	for	ADP
fcis-10369	42	7	the	the	DET
fcis-10369	42	8	minority	minority	NOUN
fcis-10369	42	9	class	class	NOUN
fcis-10369	42	10	samples	sample	NOUN
fcis-10369	42	11	,	,	PUNCT
fcis-10369	42	12	but	but	CCONJ
fcis-10369	42	13	the	the	DET
fcis-10369	42	14	overall	overall	ADJ
fcis-10369	42	15	accuracy	accuracy	NOUN
fcis-10369	42	16	was	be	AUX
fcis-10369	42	17	not	not	PART
fcis-10369	42	18	ideal	ideal	ADJ
fcis-10369	42	19	.	.	PUNCT
fcis-10369	43	1	toupas	toupas	PROPN
fcis-10369	43	2	et	et	PROPN
fcis-10369	43	3	al	al	PROPN
fcis-10369	43	4	.	.	PUNCT
fcis-10369	44	1	[	[	X
fcis-10369	44	2	16	16	NUM
fcis-10369	44	3	]	]	PUNCT
fcis-10369	44	4	proposed	propose	VERB
fcis-10369	44	5	the	the	DET
fcis-10369	44	6	combination	combination	NOUN
fcis-10369	44	7	of	of	ADP
fcis-10369	44	8	smote	smote	NOUN
fcis-10369	44	9	-	-	PUNCT
fcis-10369	44	10	enn	enn	PROPN
fcis-10369	44	11	and	and	CCONJ
fcis-10369	44	12	deep	deep	ADJ
fcis-10369	44	13	neural	neural	ADJ
fcis-10369	44	14	network	network	NOUN
fcis-10369	44	15	(	(	PUNCT
fcis-10369	44	16	dnn	dnn	PROPN
fcis-10369	44	17	)	)	PUNCT
fcis-10369	44	18	algorithm	algorithm	NOUN
fcis-10369	44	19	.	.	PUNCT
fcis-10369	45	1	however	however	ADV
fcis-10369	45	2	,	,	PUNCT
fcis-10369	45	3	they	they	PRON
fcis-10369	45	4	only	only	ADV
fcis-10369	45	5	reported	report	VERB
fcis-10369	45	6	the	the	DET
fcis-10369	45	7	results	result	NOUN
fcis-10369	45	8	of	of	ADP
fcis-10369	45	9	cross	cross	NOUN
fcis-10369	45	10	-	-	NOUN
fcis-10369	45	11	validation	validation	NOUN
fcis-10369	45	12	and	and	CCONJ
fcis-10369	45	13	did	do	AUX
fcis-10369	45	14	not	not	PART
fcis-10369	45	15	test	test	VERB
fcis-10369	45	16	on	on	ADP
fcis-10369	45	17	an	an	DET
fcis-10369	45	18	independent	independent	ADJ
fcis-10369	45	19	test	test	NOUN
fcis-10369	45	20	set	set	NOUN
fcis-10369	45	21	.	.	PUNCT
fcis-10369	46	1	in	in	ADP
fcis-10369	46	2	addressing	address	VERB
fcis-10369	46	3	the	the	DET
fcis-10369	46	4	issue	issue	NOUN
fcis-10369	46	5	of	of	ADP
fcis-10369	46	6	imbalanced	imbalanced	ADJ
fcis-10369	46	7	data	datum	NOUN
fcis-10369	46	8	,	,	PUNCT
fcis-10369	46	9	generative	generative	ADJ
fcis-10369	46	10	adversarial	adversarial	ADJ
fcis-10369	46	11	networks	network	NOUN
fcis-10369	46	12	(	(	PUNCT
fcis-10369	46	13	gans	gan	NOUN
fcis-10369	46	14	)	)	PUNCT
fcis-10369	46	15	have	have	VERB
fcis-10369	46	16	unparalleled	unparalleled	ADJ
fcis-10369	46	17	advantages	advantage	NOUN
fcis-10369	46	18	.	.	PUNCT
fcis-10369	47	1	they	they	PRON
fcis-10369	47	2	can	can	AUX
fcis-10369	47	3	learn	learn	VERB
fcis-10369	47	4	the	the	DET
fcis-10369	47	5	data	datum	NOUN
fcis-10369	47	6	distribution	distribution	NOUN
fcis-10369	47	7	and	and	CCONJ
fcis-10369	47	8	generate	generate	VERB
fcis-10369	47	9	new	new	ADJ
fcis-10369	47	10	samples	sample	NOUN
fcis-10369	47	11	from	from	ADP
fcis-10369	47	12	the	the	DET
fcis-10369	47	13	learned	learn	VERB
fcis-10369	47	14	distribution	distribution	NOUN
fcis-10369	47	15	,	,	PUNCT
fcis-10369	47	16	making	make	VERB
fcis-10369	47	17	the	the	DET
fcis-10369	47	18	generated	generate	VERB
fcis-10369	47	19	samples	sample	NOUN
fcis-10369	47	20	more	more	ADV
fcis-10369	47	21	realistic	realistic	ADJ
fcis-10369	47	22	and	and	CCONJ
fcis-10369	47	23	credible	credible	ADJ
fcis-10369	47	24	.	.	PUNCT
fcis-10369	48	1	in	in	ADP
fcis-10369	48	2	shu	shu	PROPN
fcis-10369	48	3	et	et	PROPN
fcis-10369	48	4	al	al	PROPN
fcis-10369	48	5	.	.	PUNCT
fcis-10369	49	1	[	[	X
fcis-10369	49	2	17	17	NUM
fcis-10369	49	3	]	]	PUNCT
fcis-10369	49	4	,	,	PUNCT
fcis-10369	49	5	gans	gan	NOUN
fcis-10369	49	6	were	be	AUX
fcis-10369	49	7	used	use	VERB
fcis-10369	49	8	to	to	PART
fcis-10369	49	9	generate	generate	VERB
fcis-10369	49	10	adversarial	adversarial	ADJ
fcis-10369	49	11	samples	sample	NOUN
fcis-10369	49	12	to	to	PART
fcis-10369	49	13	deceive	deceive	VERB
fcis-10369	49	14	intrusion	intrusion	NOUN
fcis-10369	49	15	detection	detection	NOUN
fcis-10369	49	16	systems	system	NOUN
fcis-10369	49	17	(	(	PUNCT
fcis-10369	49	18	ids	id	NOUN
fcis-10369	49	19	)	)	PUNCT
fcis-10369	49	20	.	.	PUNCT
fcis-10369	50	1	chen	chen	PROPN
fcis-10369	50	2	et	et	PROPN
fcis-10369	50	3	al	al	PROPN
fcis-10369	50	4	.	.	PUNCT
fcis-10369	51	1	[	[	X
fcis-10369	51	2	18	18	NUM
fcis-10369	51	3	]	]	PUNCT
fcis-10369	51	4	proposed	propose	VERB
fcis-10369	51	5	an	an	DET
fcis-10369	51	6	antiintrusion	antiintrusion	NOUN
fcis-10369	51	7	detection	detection	NOUN
fcis-10369	51	8	autoencoder	autoencoder	NOUN
fcis-10369	51	9	to	to	PART
fcis-10369	51	10	learn	learn	VERB
fcis-10369	51	11	the	the	DET
fcis-10369	51	12	distribution	distribution	NOUN
fcis-10369	51	13	of	of	ADP
fcis-10369	51	14	benign	benign	ADJ
fcis-10369	51	15	samples	sample	NOUN
fcis-10369	51	16	and	and	CCONJ
fcis-10369	51	17	generate	generate	VERB
fcis-10369	51	18	real	real	ADJ
fcis-10369	51	19	samples	sample	NOUN
fcis-10369	51	20	that	that	PRON
fcis-10369	51	21	attackers	attacker	NOUN
fcis-10369	51	22	can	can	AUX
fcis-10369	51	23	use	use	VERB
fcis-10369	51	24	to	to	PART
fcis-10369	51	25	bypass	bypass	VERB
fcis-10369	51	26	the	the	DET
fcis-10369	51	27	existing	exist	VERB
fcis-10369	51	28	ids	ids	NOUN
fcis-10369	51	29	features	feature	NOUN
fcis-10369	51	30	.	.	PUNCT
fcis-10369	52	1	although	although	SCONJ
fcis-10369	52	2	gans	gan	NOUN
fcis-10369	52	3	have	have	AUX
fcis-10369	52	4	been	be	AUX
fcis-10369	52	5	used	use	VERB
fcis-10369	52	6	to	to	PART
fcis-10369	52	7	address	address	VERB
fcis-10369	52	8	the	the	DET
fcis-10369	52	9	issue	issue	NOUN
fcis-10369	52	10	of	of	ADP
fcis-10369	52	11	class	class	NOUN
fcis-10369	52	12	imbalance	imbalance	NOUN
fcis-10369	52	13	,	,	PUNCT
fcis-10369	52	14	they	they	PRON
fcis-10369	52	15	still	still	ADV
fcis-10369	52	16	suffer	suffer	VERB
fcis-10369	52	17	from	from	ADP
fcis-10369	52	18	problems	problem	NOUN
fcis-10369	52	19	such	such	ADJ
fcis-10369	52	20	as	as	ADP
fcis-10369	52	21	mode	mode	NOUN
fcis-10369	52	22	collapse	collapse	NOUN
fcis-10369	52	23	and	and	CCONJ
fcis-10369	52	24	instability	instability	NOUN
fcis-10369	52	25	.	.	PUNCT
fcis-10369	53	1	despite	despite	SCONJ
fcis-10369	53	2	the	the	DET
fcis-10369	53	3	significant	significant	ADJ
fcis-10369	53	4	improvements	improvement	NOUN
fcis-10369	53	5	in	in	ADP
fcis-10369	53	6	detection	detection	NOUN
fcis-10369	53	7	performance	performance	NOUN
fcis-10369	53	8	achieved	achieve	VERB
fcis-10369	53	9	by	by	ADP
fcis-10369	53	10	deep	deep	ADJ
fcis-10369	53	11	learning	learning	NOUN
fcis-10369	53	12	,	,	PUNCT
fcis-10369	53	13	existing	exist	VERB
fcis-10369	53	14	intrusion	intrusion	NOUN
fcis-10369	53	15	detection	detection	NOUN
fcis-10369	53	16	methods	method	NOUN
fcis-10369	53	17	still	still	ADV
fcis-10369	53	18	have	have	VERB
fcis-10369	53	19	some	some	DET
fcis-10369	53	20	issues	issue	NOUN
fcis-10369	53	21	.	.	PUNCT
fcis-10369	54	1	(	(	PUNCT
fcis-10369	54	2	1	1	X
fcis-10369	54	3	)	)	PUNCT
fcis-10369	54	4	insufficient	insufficient	ADJ
fcis-10369	54	5	feature	feature	NOUN
fcis-10369	54	6	learning	learning	NOUN
fcis-10369	54	7	and	and	CCONJ
fcis-10369	54	8	weak	weak	ADJ
fcis-10369	54	9	model	model	NOUN
fcis-10369	54	10	representation	representation	NOUN
fcis-10369	54	11	.	.	PUNCT
fcis-10369	55	1	previous	previous	ADJ
fcis-10369	55	2	research	research	NOUN
fcis-10369	55	3	mostly	mostly	ADV
fcis-10369	55	4	focused	focus	VERB
fcis-10369	55	5	on	on	ADP
fcis-10369	55	6	a	a	DET
fcis-10369	55	7	single	single	ADJ
fcis-10369	55	8	type	type	NOUN
fcis-10369	55	9	of	of	ADP
fcis-10369	55	10	neural	neural	ADJ
fcis-10369	55	11	network	network	NOUN
fcis-10369	55	12	.	.	PUNCT
fcis-10369	56	1	cnn	cnn	PROPN
fcis-10369	56	2	-	-	PUNCT
fcis-10369	56	3	based	base	VERB
fcis-10369	56	4	models	model	NOUN
fcis-10369	56	5	can	can	AUX
fcis-10369	56	6	accurately	accurately	ADV
fcis-10369	56	7	extract	extract	VERB
fcis-10369	56	8	local	local	ADJ
fcis-10369	56	9	spatial	spatial	ADJ
fcis-10369	56	10	features	feature	NOUN
fcis-10369	56	11	but	but	CCONJ
fcis-10369	56	12	fail	fail	VERB
fcis-10369	56	13	to	to	PART
fcis-10369	56	14	learn	learn	VERB
fcis-10369	56	15	temporal	temporal	ADJ
fcis-10369	56	16	features	feature	NOUN
fcis-10369	56	17	,	,	PUNCT
fcis-10369	56	18	while	while	SCONJ
fcis-10369	56	19	rnn	rnn	NOUN
fcis-10369	56	20	-	-	PUNCT
fcis-10369	56	21	based	base	VERB
fcis-10369	56	22	models	model	NOUN
fcis-10369	56	23	can	can	AUX
fcis-10369	56	24	extract	extract	VERB
fcis-10369	56	25	temporal	temporal	ADJ
fcis-10369	56	26	features	feature	NOUN
fcis-10369	56	27	from	from	ADP
fcis-10369	56	28	data	datum	NOUN
fcis-10369	56	29	to	to	PART
fcis-10369	56	30	analyze	analyze	VERB
fcis-10369	56	31	long	long	ADJ
fcis-10369	56	32	-	-	PUNCT
fcis-10369	56	33	term	term	NOUN
fcis-10369	56	34	dependencies	dependency	NOUN
fcis-10369	56	35	but	but	CCONJ
fcis-10369	56	36	are	be	AUX
fcis-10369	56	37	not	not	PART
fcis-10369	56	38	effective	effective	ADJ
fcis-10369	56	39	in	in	ADP
fcis-10369	56	40	extracting	extract	VERB
fcis-10369	56	41	local	local	ADJ
fcis-10369	56	42	spatial	spatial	ADJ
fcis-10369	56	43	features	feature	NOUN
fcis-10369	56	44	.	.	PUNCT
fcis-10369	57	1	moreover	moreover	ADV
fcis-10369	57	2	,	,	PUNCT
fcis-10369	57	3	rnn	rnn	NOUN
fcis-10369	57	4	models	model	NOUN
fcis-10369	57	5	can	can	AUX
fcis-10369	57	6	only	only	ADV
fcis-10369	57	7	learn	learn	VERB
fcis-10369	57	8	unidirectional	unidirectional	ADJ
fcis-10369	57	9	temporal	temporal	ADJ
fcis-10369	57	10	features	feature	NOUN
fcis-10369	57	11	and	and	CCONJ
fcis-10369	57	12	do	do	AUX
fcis-10369	57	13	not	not	PART
fcis-10369	57	14	fully	fully	ADV
fcis-10369	57	15	consider	consider	VERB
fcis-10369	57	16	the	the	DET
fcis-10369	57	17	joint	joint	ADJ
fcis-10369	57	18	impact	impact	NOUN
fcis-10369	57	19	of	of	ADP
fcis-10369	57	20	preceding	precede	VERB
fcis-10369	57	21	and	and	CCONJ
fcis-10369	57	22	subsequent	subsequent	ADJ
fcis-10369	57	23	information	information	NOUN
fcis-10369	57	24	on	on	ADP
fcis-10369	57	25	the	the	DET
fcis-10369	57	26	current	current	ADJ
fcis-10369	57	27	state	state	NOUN
fcis-10369	57	28	of	of	ADP
fcis-10369	57	29	traffic	traffic	NOUN
fcis-10369	57	30	data	datum	NOUN
fcis-10369	57	31	.	.	PUNCT
fcis-10369	58	1	(	(	PUNCT
fcis-10369	58	2	2	2	NUM
fcis-10369	58	3	)	)	PUNCT
fcis-10369	58	4	class	class	NOUN
fcis-10369	58	5	imbalance	imbalance	NOUN
fcis-10369	58	6	.	.	PUNCT
fcis-10369	59	1	imbalanced	imbalanced	ADJ
fcis-10369	59	2	class	class	NOUN
fcis-10369	59	3	distribution	distribution	NOUN
fcis-10369	59	4	is	be	AUX
fcis-10369	59	5	prevalent	prevalent	ADJ
fcis-10369	59	6	in	in	ADP
fcis-10369	59	7	existing	exist	VERB
fcis-10369	59	8	publicly	publicly	ADV
fcis-10369	59	9	available	available	ADJ
fcis-10369	59	10	datasets	dataset	NOUN
fcis-10369	59	11	.	.	PUNCT
fcis-10369	60	1	however	however	ADV
fcis-10369	60	2	,	,	PUNCT
fcis-10369	60	3	most	most	ADJ
fcis-10369	60	4	researchers	researcher	NOUN
fcis-10369	60	5	have	have	AUX
fcis-10369	60	6	focused	focus	VERB
fcis-10369	60	7	on	on	ADP
fcis-10369	60	8	classifier	classifier	NOUN
fcis-10369	60	9	models	model	NOUN
fcis-10369	60	10	and	and	CCONJ
fcis-10369	60	11	paid	pay	VERB
fcis-10369	60	12	less	less	ADJ
fcis-10369	60	13	attention	attention	NOUN
fcis-10369	60	14	to	to	ADP
fcis-10369	60	15	the	the	DET
fcis-10369	60	16	impact	impact	NOUN
fcis-10369	60	17	of	of	ADP
fcis-10369	60	18	class	class	NOUN
fcis-10369	60	19	imbalance	imbalance	NOUN
fcis-10369	60	20	on	on	ADP
fcis-10369	60	21	the	the	DET
fcis-10369	60	22	performance	performance	NOUN
fcis-10369	60	23	of	of	ADP
fcis-10369	60	24	minority	minority	NOUN
fcis-10369	60	25	class	class	NOUN
fcis-10369	60	26	samples	sample	NOUN
fcis-10369	60	27	.	.	PUNCT
fcis-10369	61	1	(	(	PUNCT
fcis-10369	61	2	3	3	X
fcis-10369	61	3	)	)	PUNCT
fcis-10369	61	4	focus	focus	NOUN
fcis-10369	61	5	on	on	ADP
fcis-10369	61	6	binary	binary	ADJ
fcis-10369	61	7	classification	classification	NOUN
fcis-10369	61	8	results	result	NOUN
fcis-10369	61	9	or	or	CCONJ
fcis-10369	61	10	overall	overall	ADJ
fcis-10369	61	11	accuracy	accuracy	NOUN
fcis-10369	61	12	in	in	ADP
fcis-10369	61	13	multi	multi	ADJ
fcis-10369	61	14	-	-	ADJ
fcis-10369	61	15	class	class	ADJ
fcis-10369	61	16	scenarios	scenario	NOUN
fcis-10369	61	17	,	,	PUNCT
fcis-10369	61	18	often	often	ADV
fcis-10369	61	19	overlooking	overlook	VERB
fcis-10369	61	20	the	the	DET
fcis-10369	61	21	performance	performance	NOUN
fcis-10369	61	22	of	of	ADP
fcis-10369	61	23	individual	individual	ADJ
fcis-10369	61	24	class	class	NOUN
fcis-10369	61	25	categories	category	NOUN
fcis-10369	61	26	.	.	PUNCT
fcis-10369	62	1	to	to	PART
fcis-10369	62	2	address	address	VERB
fcis-10369	62	3	the	the	DET
fcis-10369	62	4	above	above	ADJ
fcis-10369	62	5	issues	issue	NOUN
fcis-10369	62	6	,	,	PUNCT
fcis-10369	62	7	this	this	DET
fcis-10369	62	8	paper	paper	NOUN
fcis-10369	62	9	proposes	propose	VERB
fcis-10369	62	10	a	a	DET
fcis-10369	62	11	multiscale	multiscale	ADJ
fcis-10369	62	12	feature	feature	NOUN
fcis-10369	62	13	fusion	fusion	NOUN
fcis-10369	62	14	network	network	NOUN
fcis-10369	62	15	intrusion	intrusion	NOUN
fcis-10369	62	16	detection	detection	NOUN
fcis-10369	62	17	model	model	NOUN
fcis-10369	62	18	.	.	PUNCT
fcis-10369	63	1	the	the	DET
fcis-10369	63	2	main	main	ADJ
fcis-10369	63	3	contributions	contribution	NOUN
fcis-10369	63	4	are	be	AUX
fcis-10369	63	5	as	as	SCONJ
fcis-10369	63	6	follows	follow	VERB
fcis-10369	63	7	:	:	PUNCT
fcis-10369	63	8	(	(	PUNCT
fcis-10369	63	9	1	1	X
fcis-10369	63	10	)	)	PUNCT
fcis-10369	63	11	combining	combine	VERB
fcis-10369	63	12	multi	multi	ADJ
fcis-10369	63	13	-	-	ADJ
fcis-10369	63	14	scale	scale	ADJ
fcis-10369	63	15	one	one	NUM
fcis-10369	63	16	-	-	PUNCT
fcis-10369	63	17	dimensional	dimensional	ADJ
fcis-10369	63	18	convolution	convolution	NOUN
fcis-10369	63	19	with	with	ADP
fcis-10369	63	20	bilstm	bilstm	NOUN
fcis-10369	63	21	and	and	CCONJ
fcis-10369	63	22	introducing	introduce	VERB
fcis-10369	63	23	residual	residual	ADJ
fcis-10369	63	24	connections	connection	NOUN
fcis-10369	63	25	to	to	PART
fcis-10369	63	26	fully	fully	ADV
fcis-10369	63	27	extract	extract	VERB
fcis-10369	63	28	data	data	NOUN
fcis-10369	63	29	features	feature	NOUN
fcis-10369	63	30	,	,	PUNCT
fcis-10369	63	31	thereby	thereby	ADV
fcis-10369	63	32	improving	improve	VERB
fcis-10369	63	33	the	the	DET
fcis-10369	63	34	model	model	NOUN
fcis-10369	63	35	's	's	PART
fcis-10369	63	36	representation	representation	NOUN
fcis-10369	63	37	and	and	CCONJ
fcis-10369	63	38	generalization	generalization	NOUN
fcis-10369	63	39	capabilities	capability	NOUN
fcis-10369	63	40	.	.	PUNCT
fcis-10369	64	1	(	(	PUNCT
fcis-10369	64	2	2	2	X
fcis-10369	64	3	)	)	PUNCT
fcis-10369	64	4	using	use	VERB
fcis-10369	64	5	the	the	DET
fcis-10369	64	6	wgan	wgan	ADJ
fcis-10369	64	7	-	-	PUNCT
fcis-10369	64	8	gp	gp	NOUN
fcis-10369	64	9	algorithm	algorithm	NOUN
fcis-10369	64	10	to	to	PART
fcis-10369	64	11	balance	balance	VERB
fcis-10369	64	12	the	the	DET
fcis-10369	64	13	original	original	ADJ
fcis-10369	64	14	dataset	dataset	NOUN
fcis-10369	64	15	and	and	CCONJ
fcis-10369	64	16	improve	improve	VERB
fcis-10369	64	17	the	the	DET
fcis-10369	64	18	performance	performance	NOUN
fcis-10369	64	19	of	of	ADP
fcis-10369	64	20	minority	minority	NOUN
fcis-10369	64	21	class	class	NOUN
fcis-10369	64	22	samples	sample	NOUN
fcis-10369	64	23	.	.	PUNCT
fcis-10369	65	1	(	(	PUNCT
fcis-10369	65	2	3	3	X
fcis-10369	65	3	)	)	PUNCT
fcis-10369	65	4	validating	validate	VERB
fcis-10369	65	5	the	the	DET
fcis-10369	65	6	effectiveness	effectiveness	NOUN
fcis-10369	65	7	of	of	ADP
fcis-10369	65	8	the	the	DET
fcis-10369	65	9	proposed	propose	VERB
fcis-10369	65	10	model	model	NOUN
fcis-10369	65	11	on	on	ADP
fcis-10369	65	12	the	the	DET
fcis-10369	65	13	nsl	nsl	NOUN
fcis-10369	65	14	-	-	PUNCT
fcis-10369	65	15	kdd	kdd	PROPN
fcis-10369	65	16	and	and	CCONJ
fcis-10369	65	17	unsw	unsw	PROPN
fcis-10369	65	18	-	-	PUNCT
fcis-10369	65	19	nb15	nb15	PROPN
fcis-10369	65	20	datasets	dataset	NOUN
fcis-10369	65	21	.	.	PUNCT
fcis-10369	66	1	comparing	compare	VERB
fcis-10369	66	2	the	the	DET
fcis-10369	66	3	msff	msff	NOUN
fcis-10369	66	4	model	model	NOUN
fcis-10369	66	5	with	with	ADP
fcis-10369	66	6	other	other	ADJ
fcis-10369	66	7	classical	classical	ADJ
fcis-10369	66	8	machine	machine	NOUN
fcis-10369	66	9	learning	learning	NOUN
fcis-10369	66	10	and	and	CCONJ
fcis-10369	66	11	deep	deep	ADJ
fcis-10369	66	12	learning	learning	NOUN
fcis-10369	66	13	algorithms	algorithm	NOUN
fcis-10369	66	14	in	in	ADP
fcis-10369	66	15	binary	binary	ADJ
fcis-10369	66	16	and	and	CCONJ
fcis-10369	66	17	multi	multi	ADJ
fcis-10369	66	18	-	-	ADJ
fcis-10369	66	19	class	class	ADJ
fcis-10369	66	20	experiments	experiment	NOUN
fcis-10369	66	21	.	.	PUNCT
fcis-10369	67	1	3	3	X
fcis-10369	67	2	.	.	NUM
fcis-10369	67	3	proposed	propose	VERB
fcis-10369	67	4	methodology	methodology	NOUN
fcis-10369	67	5	3.1	3.1	NUM
fcis-10369	67	6	.	.	PUNCT
fcis-10369	68	1	multi	multi	ADJ
fcis-10369	68	2	-	-	ADJ
fcis-10369	68	3	scale	scale	ADJ
fcis-10369	68	4	feature	feature	NOUN
fcis-10369	68	5	fusion	fusion	NOUN
fcis-10369	68	6	network	network	NOUN
fcis-10369	68	7	intrusion	intrusion	NOUN
fcis-10369	68	8	detection	detection	NOUN
fcis-10369	68	9	framework	framework	NOUN
fcis-10369	68	10	to	to	PART
fcis-10369	68	11	fully	fully	ADV
fcis-10369	68	12	learn	learn	VERB
fcis-10369	68	13	the	the	DET
fcis-10369	68	14	features	feature	NOUN
fcis-10369	68	15	of	of	ADP
fcis-10369	68	16	network	network	NOUN
fcis-10369	68	17	traffic	traffic	NOUN
fcis-10369	68	18	data	datum	NOUN
fcis-10369	68	19	,	,	PUNCT
fcis-10369	68	20	we	we	PRON
fcis-10369	68	21	propose	propose	VERB
fcis-10369	68	22	the	the	DET
fcis-10369	68	23	fusion	fusion	NOUN
fcis-10369	68	24	of	of	ADP
fcis-10369	68	25	multi	multi	ADJ
fcis-10369	68	26	-	-	ADJ
fcis-10369	68	27	scale	scale	ADJ
fcis-10369	68	28	one	one	NUM
fcis-10369	68	29	-	-	PUNCT
fcis-10369	68	30	dimensional	dimensional	ADJ
fcis-10369	68	31	convolution	convolution	NOUN
fcis-10369	68	32	(	(	PUNCT
fcis-10369	68	33	1d	1d	NUM
fcis-10369	68	34	-	-	PUNCT
fcis-10369	68	35	cnn	cnn	NOUN
fcis-10369	68	36	)	)	PUNCT
fcis-10369	68	37	and	and	CCONJ
fcis-10369	68	38	bilstm	bilstm	NOUN
fcis-10369	68	39	methods	method	NOUN
fcis-10369	68	40	.	.	PUNCT
fcis-10369	69	1	we	we	PRON
fcis-10369	69	2	introduce	introduce	VERB
fcis-10369	69	3	residual	residual	ADJ
fcis-10369	69	4	connections	connection	NOUN
fcis-10369	69	5	with	with	ADP
fcis-10369	69	6	identity	identity	NOUN
fcis-10369	69	7	mapping	mapping	NOUN
fcis-10369	69	8	between	between	ADP
fcis-10369	69	9	the	the	DET
fcis-10369	69	10	1dcnn	1dcnn	NUM
fcis-10369	69	11	and	and	CCONJ
fcis-10369	69	12	bilstm	bilstm	NOUN
fcis-10369	69	13	layers	layer	NOUN
fcis-10369	69	14	to	to	PART
fcis-10369	69	15	alleviate	alleviate	VERB
fcis-10369	69	16	gradient	gradient	ADJ
fcis-10369	69	17	vanishing	vanishing	NOUN
fcis-10369	69	18	and	and	CCONJ
fcis-10369	69	19	network	network	NOUN
fcis-10369	69	20	degradation	degradation	NOUN
fcis-10369	69	21	issues	issue	NOUN
fcis-10369	69	22	.	.	PUNCT
fcis-10369	70	1	the	the	DET
fcis-10369	70	2	multi	multi	ADJ
fcis-10369	70	3	-	-	ADJ
fcis-10369	70	4	scale	scale	ADJ
fcis-10369	70	5	1d	1d	NUM
fcis-10369	70	6	-	-	PUNCT
fcis-10369	70	7	cnn	cnn	NOUN
fcis-10369	70	8	uses	use	VERB
fcis-10369	70	9	filters	filter	NOUN
fcis-10369	70	10	of	of	ADP
fcis-10369	70	11	different	different	ADJ
fcis-10369	70	12	sizes	size	NOUN
fcis-10369	70	13	to	to	PART
fcis-10369	70	14	capture	capture	VERB
fcis-10369	70	15	features	feature	NOUN
fcis-10369	70	16	at	at	ADP
fcis-10369	70	17	different	different	ADJ
fcis-10369	70	18	scales	scale	NOUN
fcis-10369	70	19	.	.	PUNCT
fcis-10369	71	1	it	it	PRON
fcis-10369	71	2	can	can	AUX
fcis-10369	71	3	capture	capture	VERB
fcis-10369	71	4	both	both	CCONJ
fcis-10369	71	5	local	local	ADJ
fcis-10369	71	6	and	and	CCONJ
fcis-10369	71	7	larger	large	ADJ
fcis-10369	71	8	contextual	contextual	ADJ
fcis-10369	71	9	information	information	NOUN
fcis-10369	71	10	.	.	PUNCT
fcis-10369	72	1	the	the	DET
fcis-10369	72	2	overall	overall	ADJ
fcis-10369	72	3	framework	framework	NOUN
fcis-10369	72	4	of	of	ADP
fcis-10369	72	5	the	the	DET
fcis-10369	72	6	multi	multi	ADJ
fcis-10369	72	7	-	-	ADJ
fcis-10369	72	8	scale	scale	ADJ
fcis-10369	72	9	feature	feature	NOUN
fcis-10369	72	10	fusion	fusion	NOUN
fcis-10369	72	11	intrusion	intrusion	NOUN
fcis-10369	72	12	detection	detection	NOUN
fcis-10369	72	13	model	model	NOUN
fcis-10369	72	14	(	(	PUNCT
fcis-10369	72	15	msff	msff	NOUN
fcis-10369	72	16	-	-	PUNCT
fcis-10369	72	17	ids	id	NOUN
fcis-10369	72	18	)	)	PUNCT
fcis-10369	72	19	is	be	AUX
fcis-10369	72	20	shown	show	VERB
fcis-10369	72	21	in	in	ADP
fcis-10369	72	22	figure	figure	NOUN
fcis-10369	72	23	1	1	NUM
fcis-10369	72	24	,	,	PUNCT
fcis-10369	72	25	which	which	PRON
fcis-10369	72	26	consists	consist	VERB
fcis-10369	72	27	of	of	ADP
fcis-10369	72	28	the	the	DET
fcis-10369	72	29	data	datum	NOUN
fcis-10369	72	30	preprocessing	preprocessing	NOUN
fcis-10369	72	31	module	module	NOUN
fcis-10369	72	32	,	,	PUNCT
fcis-10369	72	33	the	the	DET
fcis-10369	72	34	mixed	mixed	ADJ
fcis-10369	72	35	sampling	sampling	NOUN
fcis-10369	72	36	module	module	NOUN
fcis-10369	72	37	,	,	PUNCT
fcis-10369	72	38	and	and	CCONJ
fcis-10369	72	39	the	the	DET
fcis-10369	72	40	multi	multi	ADJ
fcis-10369	72	41	-	-	ADJ
fcis-10369	72	42	scale	scale	ADJ
fcis-10369	72	43	feature	feature	NOUN
fcis-10369	72	44	fusion	fusion	NOUN
fcis-10369	72	45	module	module	NOUN
fcis-10369	72	46	.	.	PUNCT
fcis-10369	73	1	fig	fig	NOUN
fcis-10369	73	2	1	1	NUM
fcis-10369	73	3	.	.	PUNCT
fcis-10369	74	1	msff	msff	NOUN
fcis-10369	74	2	-	-	PUNCT
fcis-10369	74	3	framework	framework	VERB
fcis-10369	74	4	the	the	DET
fcis-10369	74	5	data	datum	NOUN
fcis-10369	74	6	preprocessing	preprocesse	VERB
fcis-10369	74	7	module	module	NOUN
fcis-10369	74	8	converts	convert	NOUN
fcis-10369	74	9	the	the	DET
fcis-10369	74	10	characterbased	characterbase	VERB
fcis-10369	74	11	features	feature	NOUN
fcis-10369	74	12	in	in	ADP
fcis-10369	74	13	the	the	DET
fcis-10369	74	14	original	original	ADJ
fcis-10369	74	15	data	datum	NOUN
fcis-10369	74	16	into	into	ADP
fcis-10369	74	17	numerical	numerical	ADJ
fcis-10369	74	18	features	feature	NOUN
fcis-10369	74	19	through	through	ADP
fcis-10369	74	20	one	one	NUM
fcis-10369	74	21	-	-	PUNCT
fcis-10369	74	22	hot	hot	ADJ
fcis-10369	74	23	encoding	encoding	NOUN
fcis-10369	74	24	and	and	CCONJ
fcis-10369	74	25	then	then	ADV
fcis-10369	74	26	applies	apply	VERB
fcis-10369	74	27	min	min	PROPN
fcis-10369	74	28	-	-	ADJ
fcis-10369	74	29	max	max	PROPN
fcis-10369	74	30	normalization	normalization	NOUN
fcis-10369	74	31	to	to	ADP
fcis-10369	74	32	the	the	DET
fcis-10369	74	33	data	datum	NOUN
fcis-10369	74	34	with	with	ADP
fcis-10369	74	35	the	the	DET
fcis-10369	74	36	same	same	ADJ
fcis-10369	74	37	attributes	attribute	NOUN
fcis-10369	74	38	.	.	PUNCT
fcis-10369	75	1	in	in	ADP
fcis-10369	75	2	the	the	DET
fcis-10369	75	3	data	data	NOUN
fcis-10369	75	4	augmentation	augmentation	NOUN
fcis-10369	75	5	module	module	NOUN
fcis-10369	75	6	,	,	PUNCT
fcis-10369	75	7	generative	generative	ADJ
fcis-10369	75	8	adversarial	adversarial	ADJ
fcis-10369	75	9	networks	network	NOUN
fcis-10369	75	10	(	(	PUNCT
fcis-10369	75	11	gans	gan	NOUN
fcis-10369	75	12	)	)	PUNCT
fcis-10369	75	13	are	be	AUX
fcis-10369	75	14	used	use	VERB
fcis-10369	75	15	to	to	PART
fcis-10369	75	16	generate	generate	VERB
fcis-10369	75	17	synthetic	synthetic	ADJ
fcis-10369	75	18	data	datum	NOUN
fcis-10369	75	19	through	through	ADP
fcis-10369	75	20	the	the	DET
fcis-10369	75	21	competition	competition	NOUN
fcis-10369	75	22	between	between	ADP
fcis-10369	75	23	a	a	DET
fcis-10369	75	24	generator	generator	NOUN
fcis-10369	75	25	and	and	CCONJ
fcis-10369	75	26	a	a	DET
fcis-10369	75	27	discriminator	discriminator	NOUN
fcis-10369	75	28	.	.	PUNCT
fcis-10369	76	1	the	the	DET
fcis-10369	76	2	generator	generator	NOUN
fcis-10369	76	3	generates	generate	VERB
fcis-10369	76	4	fake	fake	ADJ
fcis-10369	76	5	data	datum	NOUN
fcis-10369	76	6	based	base	VERB
fcis-10369	76	7	on	on	ADP
fcis-10369	76	8	input	input	NOUN
fcis-10369	76	9	random	random	ADJ
fcis-10369	76	10	noise	noise	NOUN
fcis-10369	76	11	,	,	PUNCT
fcis-10369	76	12	while	while	SCONJ
fcis-10369	76	13	the	the	DET
fcis-10369	76	14	discriminator	discriminator	NOUN
fcis-10369	76	15	distinguishes	distinguish	VERB
fcis-10369	76	16	between	between	ADP
fcis-10369	76	17	the	the	DET
fcis-10369	76	18	generated	generate	VERB
fcis-10369	76	19	fake	fake	ADJ
fcis-10369	76	20	data	datum	NOUN
fcis-10369	76	21	and	and	CCONJ
fcis-10369	76	22	real	real	ADJ
fcis-10369	76	23	samples	sample	NOUN
fcis-10369	76	24	.	.	PUNCT
fcis-10369	77	1	the	the	DET
fcis-10369	77	2	loss	loss	NOUN
fcis-10369	77	3	function	function	NOUN
fcis-10369	77	4	is	be	AUX
fcis-10369	77	5	updated	update	VERB
fcis-10369	77	6	based	base	VERB
fcis-10369	77	7	on	on	ADP
fcis-10369	77	8	the	the	DET
fcis-10369	77	9	discriminator	discriminator	NOUN
fcis-10369	77	10	's	's	PART
fcis-10369	77	11	results	result	NOUN
fcis-10369	77	12	.	.	PUNCT
fcis-10369	78	1	to	to	PART
fcis-10369	78	2	address	address	VERB
fcis-10369	78	3	the	the	DET
fcis-10369	78	4	issues	issue	NOUN
fcis-10369	78	5	of	of	ADP
fcis-10369	78	6	gradient	gradient	ADJ
fcis-10369	78	7	vanishing	vanishing	NOUN
fcis-10369	78	8	and	and	CCONJ
fcis-10369	78	9	model	model	NOUN
fcis-10369	78	10	collapse	collapse	NOUN
fcis-10369	78	11	that	that	PRON
fcis-10369	78	12	can	can	AUX
fcis-10369	78	13	occur	occur	VERB
fcis-10369	78	14	in	in	ADP
fcis-10369	78	15	gans	gan	NOUN
fcis-10369	78	16	,	,	PUNCT
fcis-10369	78	17	the	the	DET
fcis-10369	78	18	model	model	NOUN
fcis-10369	78	19	utilizes	utilize	VERB
fcis-10369	78	20	the	the	DET
fcis-10369	78	21	wasserstein	wasserstein	NOUN
fcis-10369	78	22	distance	distance	NOUN
fcis-10369	78	23	from	from	ADP
fcis-10369	78	24	wasserstein	wasserstein	PROPN
fcis-10369	78	25	gan	gan	PROPN
fcis-10369	78	26	(	(	PUNCT
fcis-10369	78	27	wgan	wgan	VERB
fcis-10369	78	28	)	)	PUNCT
fcis-10369	78	29	to	to	PART
fcis-10369	78	30	measure	measure	VERB
fcis-10369	78	31	the	the	DET
fcis-10369	78	32	distance	distance	NOUN
fcis-10369	78	33	between	between	ADP
fcis-10369	78	34	the	the	DET
fcis-10369	78	35	generated	generate	VERB
fcis-10369	78	36	distribution	distribution	NOUN
fcis-10369	78	37	and	and	CCONJ
fcis-10369	78	38	the	the	DET
fcis-10369	78	39	real	real	ADJ
fcis-10369	78	40	distribution	distribution	NOUN
fcis-10369	78	41	.	.	PUNCT
fcis-10369	79	1	this	this	PRON
fcis-10369	79	2	replaces	replace	VERB
fcis-10369	79	3	the	the	DET
fcis-10369	79	4	js	js	ADJ
fcis-10369	79	5	divergence	divergence	NOUN
fcis-10369	79	6	and	and	CCONJ
fcis-10369	79	7	kl	kl	NOUN
fcis-10369	79	8	divergence	divergence	NOUN
fcis-10369	79	9	used	use	VERB
fcis-10369	79	10	in	in	ADP
fcis-10369	79	11	gans	gan	NOUN
fcis-10369	79	12	.	.	PUNCT
fcis-10369	80	1	the	the	DET
fcis-10369	80	2	126	126	NUM
fcis-10369	80	3	wasserstein	wasserstein	NOUN
fcis-10369	80	4	distance	distance	NOUN
fcis-10369	80	5	provides	provide	VERB
fcis-10369	80	6	a	a	DET
fcis-10369	80	7	smooth	smooth	ADJ
fcis-10369	80	8	representation	representation	NOUN
fcis-10369	80	9	of	of	ADP
fcis-10369	80	10	the	the	DET
fcis-10369	80	11	proximity	proximity	NOUN
fcis-10369	80	12	between	between	ADP
fcis-10369	80	13	the	the	DET
fcis-10369	80	14	two	two	NUM
fcis-10369	80	15	distributions	distribution	NOUN
fcis-10369	80	16	.	.	PUNCT
fcis-10369	81	1	additionally	additionally	ADV
fcis-10369	81	2	,	,	PUNCT
fcis-10369	81	3	to	to	PART
fcis-10369	81	4	overcome	overcome	VERB
fcis-10369	81	5	the	the	DET
fcis-10369	81	6	limitations	limitation	NOUN
fcis-10369	81	7	and	and	CCONJ
fcis-10369	81	8	difficulties	difficulty	NOUN
fcis-10369	81	9	of	of	ADP
fcis-10369	81	10	weight	weight	NOUN
fcis-10369	81	11	clipping	clip	VERB
fcis-10369	81	12	in	in	ADP
fcis-10369	81	13	wgan	wgan	ADJ
fcis-10369	81	14	,	,	PUNCT
fcis-10369	81	15	wgan	wgan	ADJ
fcis-10369	81	16	-	-	PUNCT
fcis-10369	81	17	gp	gp	NOUN
fcis-10369	81	18	introduces	introduce	NOUN
fcis-10369	81	19	gradient	gradient	NOUN
fcis-10369	81	20	penalty	penalty	NOUN
fcis-10369	81	21	techniques	technique	NOUN
fcis-10369	81	22	to	to	PART
fcis-10369	81	23	train	train	VERB
fcis-10369	81	24	the	the	DET
fcis-10369	81	25	generator	generator	NOUN
fcis-10369	81	26	and	and	CCONJ
fcis-10369	81	27	discriminator	discriminator	NOUN
fcis-10369	81	28	more	more	ADV
fcis-10369	81	29	stably	stably	ADV
fcis-10369	81	30	,	,	PUNCT
fcis-10369	81	31	thereby	thereby	ADV
fcis-10369	81	32	improving	improve	VERB
fcis-10369	81	33	the	the	DET
fcis-10369	81	34	performance	performance	NOUN
fcis-10369	81	35	of	of	ADP
fcis-10369	81	36	the	the	DET
fcis-10369	81	37	generative	generative	ADJ
fcis-10369	81	38	adversarial	adversarial	ADJ
fcis-10369	81	39	network	network	NOUN
fcis-10369	81	40	.	.	PUNCT
fcis-10369	82	1	the	the	DET
fcis-10369	82	2	multi	multi	ADJ
fcis-10369	82	3	-	-	ADJ
fcis-10369	82	4	scale	scale	ADJ
fcis-10369	82	5	feature	feature	NOUN
fcis-10369	82	6	fusion	fusion	NOUN
fcis-10369	82	7	module	module	NOUN
fcis-10369	82	8	focuses	focus	VERB
fcis-10369	82	9	on	on	ADP
fcis-10369	82	10	different	different	ADJ
fcis-10369	82	11	types	type	NOUN
fcis-10369	82	12	of	of	ADP
fcis-10369	82	13	features	feature	NOUN
fcis-10369	82	14	through	through	ADP
fcis-10369	82	15	multiple	multiple	ADJ
fcis-10369	82	16	one	one	NUM
fcis-10369	82	17	-	-	PUNCT
fcis-10369	82	18	dimensional	dimensional	ADJ
fcis-10369	82	19	convolutional	convolutional	ADJ
fcis-10369	82	20	layers	layer	NOUN
fcis-10369	82	21	with	with	ADP
fcis-10369	82	22	different	different	ADJ
fcis-10369	82	23	filters	filter	NOUN
fcis-10369	82	24	.	.	PUNCT
fcis-10369	83	1	network	network	NOUN
fcis-10369	83	2	traffic	traffic	NOUN
fcis-10369	83	3	data	datum	NOUN
fcis-10369	83	4	may	may	AUX
fcis-10369	83	5	contain	contain	VERB
fcis-10369	83	6	various	various	ADJ
fcis-10369	83	7	types	type	NOUN
fcis-10369	83	8	of	of	ADP
fcis-10369	83	9	features	feature	NOUN
fcis-10369	83	10	such	such	ADJ
fcis-10369	83	11	as	as	ADP
fcis-10369	83	12	source	source	NOUN
fcis-10369	83	13	addresses	address	NOUN
fcis-10369	83	14	,	,	PUNCT
fcis-10369	83	15	destination	destination	NOUN
fcis-10369	83	16	addresses	address	NOUN
fcis-10369	83	17	,	,	PUNCT
fcis-10369	83	18	protocol	protocol	NOUN
fcis-10369	83	19	types	type	NOUN
fcis-10369	83	20	,	,	PUNCT
fcis-10369	83	21	and	and	CCONJ
fcis-10369	83	22	port	port	NOUN
fcis-10369	83	23	numbers	number	NOUN
fcis-10369	83	24	.	.	PUNCT
fcis-10369	84	1	by	by	ADP
fcis-10369	84	2	designing	design	VERB
fcis-10369	84	3	filters	filter	NOUN
fcis-10369	84	4	of	of	ADP
fcis-10369	84	5	different	different	ADJ
fcis-10369	84	6	sizes	size	NOUN
fcis-10369	84	7	,	,	PUNCT
fcis-10369	84	8	the	the	DET
fcis-10369	84	9	module	module	NOUN
fcis-10369	84	10	specifically	specifically	ADV
fcis-10369	84	11	attends	attend	VERB
fcis-10369	84	12	to	to	ADP
fcis-10369	84	13	a	a	DET
fcis-10369	84	14	certain	certain	ADJ
fcis-10369	84	15	type	type	NOUN
fcis-10369	84	16	of	of	ADP
fcis-10369	84	17	feature	feature	NOUN
fcis-10369	84	18	and	and	CCONJ
fcis-10369	84	19	extracts	extract	VERB
fcis-10369	84	20	its	its	PRON
fcis-10369	84	21	representation	representation	NOUN
fcis-10369	84	22	.	.	PUNCT
fcis-10369	85	1	this	this	DET
fcis-10369	85	2	combination	combination	NOUN
fcis-10369	85	3	enhances	enhance	VERB
fcis-10369	85	4	the	the	DET
fcis-10369	85	5	model	model	NOUN
fcis-10369	85	6	's	's	PART
fcis-10369	85	7	ability	ability	NOUN
fcis-10369	85	8	to	to	PART
fcis-10369	85	9	learn	learn	VERB
fcis-10369	85	10	different	different	ADJ
fcis-10369	85	11	types	type	NOUN
fcis-10369	85	12	of	of	ADP
fcis-10369	85	13	features	feature	NOUN
fcis-10369	85	14	and	and	CCONJ
fcis-10369	85	15	captures	capture	VERB
fcis-10369	85	16	diverse	diverse	ADJ
fcis-10369	85	17	feature	feature	NOUN
fcis-10369	85	18	patterns	pattern	NOUN
fcis-10369	85	19	in	in	ADP
fcis-10369	85	20	network	network	NOUN
fcis-10369	85	21	traffic	traffic	NOUN
fcis-10369	85	22	data	datum	NOUN
fcis-10369	85	23	.	.	PUNCT
fcis-10369	86	1	then	then	ADV
fcis-10369	86	2	,	,	PUNCT
fcis-10369	86	3	two	two	NUM
fcis-10369	86	4	layers	layer	NOUN
fcis-10369	86	5	of	of	ADP
fcis-10369	86	6	bilstm	bilstm	NOUN
fcis-10369	86	7	are	be	AUX
fcis-10369	86	8	used	use	VERB
fcis-10369	86	9	to	to	PART
fcis-10369	86	10	extract	extract	VERB
fcis-10369	86	11	temporal	temporal	ADJ
fcis-10369	86	12	patterns	pattern	NOUN
fcis-10369	86	13	and	and	CCONJ
fcis-10369	86	14	global	global	ADJ
fcis-10369	86	15	features	feature	NOUN
fcis-10369	86	16	from	from	ADP
fcis-10369	86	17	the	the	DET
fcis-10369	86	18	network	network	NOUN
fcis-10369	86	19	traffic	traffic	NOUN
fcis-10369	86	20	data	datum	NOUN
fcis-10369	86	21	.	.	PUNCT
fcis-10369	87	1	finally	finally	ADV
fcis-10369	87	2	,	,	PUNCT
fcis-10369	87	3	the	the	DET
fcis-10369	87	4	fused	fuse	VERB
fcis-10369	87	5	features	feature	NOUN
fcis-10369	87	6	are	be	AUX
fcis-10369	87	7	passed	pass	VERB
fcis-10369	87	8	through	through	ADP
fcis-10369	87	9	a	a	DET
fcis-10369	87	10	softmax	softmax	NOUN
fcis-10369	87	11	classifier	classifier	NOUN
fcis-10369	87	12	to	to	PART
fcis-10369	87	13	obtain	obtain	VERB
fcis-10369	87	14	classification	classification	NOUN
fcis-10369	87	15	results	result	NOUN
fcis-10369	87	16	.	.	PUNCT
fcis-10369	88	1	the	the	DET
fcis-10369	88	2	implementation	implementation	NOUN
fcis-10369	88	3	process	process	NOUN
fcis-10369	88	4	of	of	ADP
fcis-10369	88	5	network	network	NOUN
fcis-10369	88	6	intrusion	intrusion	NOUN
fcis-10369	88	7	detection	detection	NOUN
fcis-10369	88	8	based	base	VERB
fcis-10369	88	9	on	on	ADP
fcis-10369	88	10	the	the	DET
fcis-10369	88	11	multi	multi	ADJ
fcis-10369	88	12	-	-	ADJ
fcis-10369	88	13	scale	scale	ADJ
fcis-10369	88	14	feature	feature	NOUN
fcis-10369	88	15	fusion	fusion	NOUN
fcis-10369	88	16	(	(	PUNCT
fcis-10369	88	17	msff	msff	NOUN
fcis-10369	88	18	)	)	PUNCT
fcis-10369	88	19	is	be	AUX
fcis-10369	88	20	shown	show	VERB
fcis-10369	88	21	in	in	ADP
fcis-10369	88	22	algorithm	algorithm	NOUN
fcis-10369	88	23	1	1	NUM
fcis-10369	88	24	.	.	PUNCT
fcis-10369	88	25	algorithm	algorithm	NOUN
fcis-10369	88	26	1	1	NUM
fcis-10369	88	27	:	:	PUNCT
fcis-10369	88	28	network	network	NOUN
fcis-10369	88	29	intrusion	intrusion	NOUN
fcis-10369	88	30	detection	detection	NOUN
fcis-10369	88	31	implementation	implementation	NOUN
fcis-10369	88	32	based	base	VERB
fcis-10369	88	33	on	on	ADP
fcis-10369	88	34	msff	msff	ADJ
fcis-10369	88	35	input	input	NOUN
fcis-10369	88	36	:	:	PUNCT
fcis-10369	88	37	original	original	ADJ
fcis-10369	88	38	dataset	dataset	NOUN
fcis-10369	88	39	output	output	NOUN
fcis-10369	88	40	:	:	PUNCT
fcis-10369	88	41	classification	classification	NOUN
fcis-10369	88	42	results	result	NOUN
fcis-10369	88	43	step	step	VERB
fcis-10369	88	44	1	1	NUM
fcis-10369	88	45	:	:	PUNCT
fcis-10369	88	46	data	datum	NOUN
fcis-10369	88	47	preprocessing	preprocesse	VERB
fcis-10369	88	48	1	1	NUM
fcis-10369	88	49	.	.	NUM
fcis-10369	88	50	encode	encode	VERB
fcis-10369	88	51	and	and	CCONJ
fcis-10369	88	52	convert	convert	VERB
fcis-10369	88	53	the	the	DET
fcis-10369	88	54	original	original	ADJ
fcis-10369	88	55	dataset	dataset	NOUN
fcis-10369	88	56	into	into	ADP
fcis-10369	88	57	numerical	numerical	ADJ
fcis-10369	88	58	values	value	NOUN
fcis-10369	88	59	.	.	PUNCT
fcis-10369	89	1	2	2	X
fcis-10369	89	2	.	.	X
fcis-10369	89	3	normalize	normalize	VERB
fcis-10369	89	4	the	the	DET
fcis-10369	89	5	dataset	dataset	NOUN
fcis-10369	89	6	after	after	ADP
fcis-10369	89	7	numerical	numerical	PROPN
fcis-10369	89	8	conversion	conversion	PROPN
fcis-10369	89	9	.	.	PUNCT
fcis-10369	90	1	step	step	NOUN
fcis-10369	90	2	2	2	NUM
fcis-10369	90	3	:	:	PUNCT
fcis-10369	90	4	construct	construct	VERB
fcis-10369	90	5	a	a	DET
fcis-10369	90	6	balanced	balanced	ADJ
fcis-10369	90	7	dataset	dataset	NOUN
fcis-10369	90	8	3	3	NUM
fcis-10369	90	9	.	.	PUNCT
fcis-10369	90	10	use	use	VERB
fcis-10369	90	11	the	the	DET
fcis-10369	90	12	smote	smote	NOUN
fcis-10369	90	13	-	-	PUNCT
fcis-10369	90	14	tomek	tomek	PROPN
fcis-10369	90	15	algorithm	algorithm	NOUN
fcis-10369	90	16	to	to	PART
fcis-10369	90	17	perform	perform	VERB
fcis-10369	90	18	mixed	mixed	ADJ
fcis-10369	90	19	sampling	sampling	NOUN
fcis-10369	90	20	on	on	ADP
fcis-10369	90	21	the	the	DET
fcis-10369	90	22	training	training	NOUN
fcis-10369	90	23	set	set	NOUN
fcis-10369	90	24	and	and	CCONJ
fcis-10369	90	25	reconstruct	reconstruct	VERB
fcis-10369	90	26	the	the	DET
fcis-10369	90	27	unprocessed	unprocessed	ADJ
fcis-10369	90	28	test	test	NOUN
fcis-10369	90	29	set	set	VERB
fcis-10369	90	30	into	into	ADP
fcis-10369	90	31	a	a	DET
fcis-10369	90	32	balanced	balanced	ADJ
fcis-10369	90	33	dataset	dataset	NOUN
fcis-10369	90	34	.	.	PUNCT
fcis-10369	91	1	step	step	NOUN
fcis-10369	91	2	3	3	NUM
fcis-10369	91	3	:	:	PUNCT
fcis-10369	91	4	build	build	VERB
fcis-10369	91	5	the	the	DET
fcis-10369	91	6	model	model	NOUN
fcis-10369	91	7	4	4	NUM
fcis-10369	91	8	.	.	PUNCT
fcis-10369	91	9	use	use	VERB
fcis-10369	91	10	multi	multi	ADJ
fcis-10369	91	11	-	-	ADJ
fcis-10369	91	12	scale	scale	ADJ
fcis-10369	91	13	convolution	convolution	NOUN
fcis-10369	91	14	to	to	PART
fcis-10369	91	15	extract	extract	VERB
fcis-10369	91	16	local	local	ADJ
fcis-10369	91	17	spatial	spatial	ADJ
fcis-10369	91	18	features	feature	NOUN
fcis-10369	91	19	and	and	CCONJ
fcis-10369	91	20	implement	implement	VERB
fcis-10369	91	21	parameter	parameter	NOUN
fcis-10369	91	22	sharing	sharing	NOUN
fcis-10369	91	23	.	.	PUNCT
fcis-10369	92	1	5	5	X
fcis-10369	92	2	.	.	X
fcis-10369	92	3	add	add	VERB
fcis-10369	92	4	pooling	pool	VERB
fcis-10369	92	5	layers	layer	NOUN
fcis-10369	92	6	to	to	PART
fcis-10369	92	7	prevent	prevent	VERB
fcis-10369	92	8	overfitting	overfitting	NOUN
fcis-10369	92	9	.	.	PUNCT
fcis-10369	93	1	6	6	X
fcis-10369	93	2	.	.	X
fcis-10369	93	3	add	add	VERB
fcis-10369	93	4	batch	batch	NOUN
fcis-10369	93	5	normalization	normalization	NOUN
fcis-10369	93	6	layers	layer	NOUN
fcis-10369	93	7	to	to	PART
fcis-10369	93	8	normalize	normalize	VERB
fcis-10369	93	9	the	the	DET
fcis-10369	93	10	parameters	parameter	NOUN
fcis-10369	93	11	of	of	ADP
fcis-10369	93	12	the	the	DET
fcis-10369	93	13	previous	previous	ADJ
fcis-10369	93	14	layer	layer	NOUN
fcis-10369	93	15	and	and	CCONJ
fcis-10369	93	16	prevent	prevent	VERB
fcis-10369	93	17	training	training	NOUN
fcis-10369	93	18	slowdown	slowdown	NOUN
fcis-10369	93	19	.	.	PUNCT
fcis-10369	94	1	7	7	X
fcis-10369	94	2	.	.	X
fcis-10369	94	3	add	add	VERB
fcis-10369	94	4	bilstm	bilstm	NOUN
fcis-10369	94	5	layers	layer	NOUN
fcis-10369	94	6	to	to	PART
fcis-10369	94	7	learn	learn	VERB
fcis-10369	94	8	from	from	ADP
fcis-10369	94	9	forward	forward	ADV
fcis-10369	94	10	and	and	CCONJ
fcis-10369	94	11	backward	backward	ADJ
fcis-10369	94	12	temporal	temporal	ADJ
fcis-10369	94	13	sequences	sequence	NOUN
fcis-10369	94	14	.	.	PUNCT
fcis-10369	95	1	8	8	X
fcis-10369	95	2	.	.	X
fcis-10369	95	3	introduce	introduce	VERB
fcis-10369	95	4	residual	residual	ADJ
fcis-10369	95	5	connections	connection	NOUN
fcis-10369	95	6	with	with	ADP
fcis-10369	95	7	identity	identity	NOUN
fcis-10369	95	8	mappings	mapping	NOUN
fcis-10369	95	9	between	between	ADP
fcis-10369	95	10	the	the	DET
fcis-10369	95	11	input	input	NOUN
fcis-10369	95	12	layer	layer	NOUN
fcis-10369	95	13	and	and	CCONJ
fcis-10369	95	14	bilstm	bilstm	NOUN
fcis-10369	95	15	layers	layer	NOUN
fcis-10369	95	16	.	.	PUNCT
fcis-10369	96	1	9	9	X
fcis-10369	96	2	.	.	X
fcis-10369	96	3	repeat	repeat	NOUN
fcis-10369	96	4	steps	step	NOUN
fcis-10369	96	5	5	5	NUM
fcis-10369	96	6	-	-	SYM
fcis-10369	96	7	7	7	NUM
fcis-10369	96	8	and	and	CCONJ
fcis-10369	96	9	add	add	VERB
fcis-10369	96	10	dropout	dropout	NOUN
fcis-10369	96	11	layers	layer	NOUN
fcis-10369	96	12	and	and	CCONJ
fcis-10369	96	13	fully	fully	ADV
fcis-10369	96	14	connected	connected	ADJ
fcis-10369	96	15	layers	layer	NOUN
fcis-10369	96	16	.	.	PUNCT
fcis-10369	97	1	10	10	NUM
fcis-10369	97	2	.	.	X
fcis-10369	97	3	use	use	VERB
fcis-10369	97	4	a	a	DET
fcis-10369	97	5	softmax	softmax	NOUN
fcis-10369	97	6	classifier	classifier	NOUN
fcis-10369	97	7	for	for	ADP
fcis-10369	97	8	classification	classification	NOUN
fcis-10369	97	9	.	.	PUNCT
fcis-10369	98	1	step	step	NOUN
fcis-10369	98	2	4	4	NUM
fcis-10369	98	3	:	:	PUNCT
fcis-10369	98	4	train	train	VERB
fcis-10369	98	5	the	the	DET
fcis-10369	98	6	model	model	NOUN
fcis-10369	98	7	11.set	11.set	PROPN
fcis-10369	98	8	experimental	experimental	ADJ
fcis-10369	98	9	hyperparameters	hyperparameter	NOUN
fcis-10369	98	10	and	and	CCONJ
fcis-10369	98	11	network	network	NOUN
fcis-10369	98	12	structure	structure	NOUN
fcis-10369	98	13	parameters	parameter	NOUN
fcis-10369	98	14	as	as	SCONJ
fcis-10369	98	15	described	describe	VERB
fcis-10369	98	16	in	in	ADP
fcis-10369	98	17	section	section	NOUN
fcis-10369	98	18	3.3	3.3	NUM
fcis-10369	98	19	.	.	PUNCT
fcis-10369	99	1	step	step	NOUN
fcis-10369	99	2	5	5	NUM
fcis-10369	99	3	:	:	PUNCT
fcis-10369	99	4	test	test	VERB
fcis-10369	99	5	the	the	DET
fcis-10369	99	6	model	model	NOUN
fcis-10369	99	7	12	12	NUM
fcis-10369	99	8	.	.	PUNCT
fcis-10369	100	1	feed	feed	VERB
fcis-10369	100	2	the	the	DET
fcis-10369	100	3	test	test	NOUN
fcis-10369	100	4	set	set	VERB
fcis-10369	100	5	into	into	ADP
fcis-10369	100	6	the	the	DET
fcis-10369	100	7	trained	train	VERB
fcis-10369	100	8	model	model	NOUN
fcis-10369	100	9	and	and	CCONJ
fcis-10369	100	10	obtain	obtain	VERB
fcis-10369	100	11	the	the	DET
fcis-10369	100	12	classification	classification	NOUN
fcis-10369	100	13	results	result	NOUN
fcis-10369	100	14	.	.	PUNCT
fcis-10369	101	1	3.2	3.2	NUM
fcis-10369	101	2	.	.	PUNCT
fcis-10369	102	1	multi	multi	ADJ
fcis-10369	102	2	-	-	ADJ
fcis-10369	102	3	scale	scale	ADJ
fcis-10369	102	4	feature	feature	NOUN
fcis-10369	102	5	fusion	fusion	NOUN
fcis-10369	102	6	network	network	NOUN
fcis-10369	102	7	the	the	DET
fcis-10369	102	8	multi	multi	ADJ
fcis-10369	102	9	-	-	ADJ
fcis-10369	102	10	scale	scale	ADJ
fcis-10369	102	11	feature	feature	NOUN
fcis-10369	102	12	fusion	fusion	NOUN
fcis-10369	102	13	(	(	PUNCT
fcis-10369	102	14	msff	msff	NOUN
fcis-10369	102	15	)	)	PUNCT
fcis-10369	102	16	network	network	NOUN
fcis-10369	102	17	module	module	NOUN
fcis-10369	102	18	is	be	AUX
fcis-10369	102	19	the	the	DET
fcis-10369	102	20	core	core	NOUN
fcis-10369	102	21	component	component	NOUN
fcis-10369	102	22	of	of	ADP
fcis-10369	102	23	msff	msff	NOUN
fcis-10369	102	24	-	-	PUNCT
fcis-10369	102	25	ids	ids	NOUN
fcis-10369	102	26	.	.	PUNCT
fcis-10369	103	1	in	in	ADP
fcis-10369	103	2	this	this	DET
fcis-10369	103	3	module	module	NOUN
fcis-10369	103	4	,	,	PUNCT
fcis-10369	103	5	the	the	DET
fcis-10369	103	6	multiscale	multiscale	ADJ
fcis-10369	103	7	1d	1d	NUM
fcis-10369	103	8	convolution	convolution	NOUN
fcis-10369	103	9	is	be	AUX
fcis-10369	103	10	utilized	utilize	VERB
fcis-10369	103	11	to	to	PART
fcis-10369	103	12	consider	consider	VERB
fcis-10369	103	13	features	feature	NOUN
fcis-10369	103	14	of	of	ADP
fcis-10369	103	15	different	different	ADJ
fcis-10369	103	16	scales	scale	NOUN
fcis-10369	103	17	,	,	PUNCT
fcis-10369	103	18	enriching	enrich	VERB
fcis-10369	103	19	the	the	DET
fcis-10369	103	20	model	model	NOUN
fcis-10369	103	21	's	's	PART
fcis-10369	103	22	representation	representation	NOUN
fcis-10369	103	23	capability	capability	NOUN
fcis-10369	103	24	of	of	ADP
fcis-10369	103	25	input	input	NOUN
fcis-10369	103	26	data	datum	NOUN
fcis-10369	103	27	.	.	PUNCT
fcis-10369	104	1	by	by	ADP
fcis-10369	104	2	introducing	introduce	VERB
fcis-10369	104	3	filters	filter	NOUN
fcis-10369	104	4	of	of	ADP
fcis-10369	104	5	different	different	ADJ
fcis-10369	104	6	scales	scale	NOUN
fcis-10369	104	7	in	in	ADP
fcis-10369	104	8	the	the	DET
fcis-10369	104	9	convolutional	convolutional	ADJ
fcis-10369	104	10	layer	layer	NOUN
fcis-10369	104	11	,	,	PUNCT
fcis-10369	104	12	a	a	DET
fcis-10369	104	13	more	more	ADV
fcis-10369	104	14	comprehensive	comprehensive	ADJ
fcis-10369	104	15	feature	feature	NOUN
fcis-10369	104	16	representation	representation	NOUN
fcis-10369	104	17	is	be	AUX
fcis-10369	104	18	provided	provide	VERB
fcis-10369	104	19	,	,	PUNCT
fcis-10369	104	20	enhancing	enhance	VERB
fcis-10369	104	21	the	the	DET
fcis-10369	104	22	model	model	NOUN
fcis-10369	104	23	's	's	PART
fcis-10369	104	24	expression	expression	NOUN
fcis-10369	104	25	and	and	CCONJ
fcis-10369	104	26	generalization	generalization	NOUN
fcis-10369	104	27	ability	ability	NOUN
fcis-10369	104	28	.	.	PUNCT
fcis-10369	105	1	the	the	DET
fcis-10369	105	2	residual	residual	ADJ
fcis-10369	105	3	connection	connection	NOUN
fcis-10369	105	4	,	,	PUNCT
fcis-10369	105	5	which	which	PRON
fcis-10369	105	6	passes	pass	VERB
fcis-10369	105	7	the	the	DET
fcis-10369	105	8	original	original	ADJ
fcis-10369	105	9	feature	feature	NOUN
fcis-10369	105	10	information	information	NOUN
fcis-10369	105	11	directly	directly	ADV
fcis-10369	105	12	between	between	ADP
fcis-10369	105	13	the	the	DET
fcis-10369	105	14	input	input	NOUN
fcis-10369	105	15	layer	layer	NOUN
fcis-10369	105	16	and	and	CCONJ
fcis-10369	105	17	bilstm	bilstm	NOUN
fcis-10369	105	18	,	,	PUNCT
fcis-10369	105	19	improves	improve	VERB
fcis-10369	105	20	feature	feature	NOUN
fcis-10369	105	21	expression	expression	NOUN
fcis-10369	105	22	and	and	CCONJ
fcis-10369	105	23	model	model	NOUN
fcis-10369	105	24	convergence	convergence	NOUN
fcis-10369	105	25	speed	speed	NOUN
fcis-10369	105	26	,	,	PUNCT
fcis-10369	105	27	while	while	SCONJ
fcis-10369	105	28	addressing	address	VERB
fcis-10369	105	29	the	the	DET
fcis-10369	105	30	issue	issue	NOUN
fcis-10369	105	31	of	of	ADP
fcis-10369	105	32	gradient	gradient	ADJ
fcis-10369	105	33	vanishing	vanishing	NOUN
fcis-10369	105	34	or	or	CCONJ
fcis-10369	105	35	exploding	explode	VERB
fcis-10369	105	36	commonly	commonly	ADV
fcis-10369	105	37	encountered	encounter	VERB
fcis-10369	105	38	in	in	ADP
fcis-10369	105	39	traditional	traditional	ADJ
fcis-10369	105	40	deep	deep	ADJ
fcis-10369	105	41	neural	neural	ADJ
fcis-10369	105	42	networks	network	NOUN
fcis-10369	105	43	.	.	PUNCT
fcis-10369	106	1	the	the	DET
fcis-10369	106	2	structure	structure	NOUN
fcis-10369	106	3	of	of	ADP
fcis-10369	106	4	the	the	DET
fcis-10369	106	5	residual	residual	ADJ
fcis-10369	106	6	module	module	NOUN
fcis-10369	106	7	is	be	AUX
fcis-10369	106	8	shown	show	VERB
fcis-10369	106	9	in	in	ADP
fcis-10369	106	10	figure	figure	NOUN
fcis-10369	106	11	2	2	NUM
fcis-10369	106	12	.	.	PUNCT
fcis-10369	107	1	the	the	DET
fcis-10369	107	2	following	follow	VERB
fcis-10369	107	3	sections	section	NOUN
fcis-10369	107	4	will	will	AUX
fcis-10369	107	5	provide	provide	VERB
fcis-10369	107	6	an	an	DET
fcis-10369	107	7	introduction	introduction	NOUN
fcis-10369	107	8	to	to	ADP
fcis-10369	107	9	the	the	DET
fcis-10369	107	10	multi	multi	ADJ
fcis-10369	107	11	-	-	ADJ
fcis-10369	107	12	scale	scale	ADJ
fcis-10369	107	13	convolutional	convolutional	ADJ
fcis-10369	107	14	layer	layer	NOUN
fcis-10369	107	15	and	and	CCONJ
fcis-10369	107	16	the	the	DET
fcis-10369	107	17	identity	identity	NOUN
fcis-10369	107	18	mapping	mapping	NOUN
fcis-10369	107	19	.	.	PUNCT
fcis-10369	108	1	fig	fig	NOUN
fcis-10369	108	2	2	2	NUM
fcis-10369	108	3	.	.	PUNCT
fcis-10369	108	4	residual	residual	ADJ
fcis-10369	108	5	module	module	NOUN
fcis-10369	108	6	structure	structure	NOUN
fcis-10369	108	7	3.2.1	3.2.1	NUM
fcis-10369	108	8	.	.	PUNCT
fcis-10369	109	1	multi	multi	ADJ
fcis-10369	109	2	-	-	ADJ
fcis-10369	109	3	scale	scale	ADJ
fcis-10369	109	4	convolutional	convolutional	ADJ
fcis-10369	109	5	layer	layer	NOUN
fcis-10369	109	6	one	one	NUM
fcis-10369	109	7	-	-	PUNCT
fcis-10369	109	8	dimensional	dimensional	ADJ
fcis-10369	109	9	convolution	convolution	NOUN
fcis-10369	109	10	primarily	primarily	ADV
fcis-10369	109	11	focuses	focus	VERB
fcis-10369	109	12	on	on	ADP
fcis-10369	109	13	local	local	ADJ
fcis-10369	109	14	details	detail	NOUN
fcis-10369	109	15	,	,	PUNCT
fcis-10369	109	16	while	while	SCONJ
fcis-10369	109	17	multi	multi	ADJ
fcis-10369	109	18	-	-	ADJ
fcis-10369	109	19	scale	scale	ADJ
fcis-10369	109	20	one	one	NUM
fcis-10369	109	21	-	-	PUNCT
fcis-10369	109	22	dimensional	dimensional	ADJ
fcis-10369	109	23	convolution	convolution	NOUN
fcis-10369	109	24	can	can	AUX
fcis-10369	109	25	provide	provide	VERB
fcis-10369	109	26	a	a	DET
fcis-10369	109	27	more	more	ADV
fcis-10369	109	28	comprehensive	comprehensive	ADJ
fcis-10369	109	29	description	description	NOUN
fcis-10369	109	30	of	of	ADP
fcis-10369	109	31	the	the	DET
fcis-10369	109	32	relationships	relationship	NOUN
fcis-10369	109	33	between	between	ADP
fcis-10369	109	34	global	global	ADJ
fcis-10369	109	35	and	and	CCONJ
fcis-10369	109	36	local	local	ADJ
fcis-10369	109	37	features	feature	NOUN
fcis-10369	109	38	.	.	PUNCT
fcis-10369	110	1	to	to	PART
fcis-10369	110	2	effectively	effectively	ADV
fcis-10369	110	3	extract	extract	VERB
fcis-10369	110	4	the	the	DET
fcis-10369	110	5	local	local	ADJ
fcis-10369	110	6	spatial	spatial	ADJ
fcis-10369	110	7	features	feature	NOUN
fcis-10369	110	8	of	of	ADP
fcis-10369	110	9	network	network	NOUN
fcis-10369	110	10	traffic	traffic	NOUN
fcis-10369	110	11	data	datum	NOUN
fcis-10369	110	12	,	,	PUNCT
fcis-10369	110	13	this	this	DET
fcis-10369	110	14	study	study	NOUN
fcis-10369	110	15	adopts	adopt	VERB
fcis-10369	110	16	multiple	multiple	ADJ
fcis-10369	110	17	filters	filter	NOUN
fcis-10369	110	18	of	of	ADP
fcis-10369	110	19	different	different	ADJ
fcis-10369	110	20	scales	scale	NOUN
fcis-10369	110	21	(	(	PUNCT
fcis-10369	110	22	16	16	NUM
fcis-10369	110	23	,	,	PUNCT
fcis-10369	110	24	32	32	NUM
fcis-10369	110	25	,	,	PUNCT
fcis-10369	110	26	64	64	NUM
fcis-10369	110	27	)	)	PUNCT
fcis-10369	110	28	to	to	PART
fcis-10369	110	29	capture	capture	VERB
fcis-10369	110	30	features	feature	NOUN
fcis-10369	110	31	at	at	ADP
fcis-10369	110	32	different	different	ADJ
fcis-10369	110	33	levels	level	NOUN
fcis-10369	110	34	.	.	PUNCT
fcis-10369	111	1	in	in	ADP
fcis-10369	111	2	multi	multi	ADJ
fcis-10369	111	3	-	-	ADJ
fcis-10369	111	4	scale	scale	ADJ
fcis-10369	111	5	convolution	convolution	NOUN
fcis-10369	111	6	,	,	PUNCT
fcis-10369	111	7	both	both	CCONJ
fcis-10369	111	8	the	the	DET
fcis-10369	111	9	concatenate	concatenate	NOUN
fcis-10369	111	10	function	function	NOUN
fcis-10369	111	11	and	and	CCONJ
fcis-10369	111	12	the	the	DET
fcis-10369	111	13	add	add	ADJ
fcis-10369	111	14	function	function	NOUN
fcis-10369	111	15	can	can	AUX
fcis-10369	111	16	be	be	AUX
fcis-10369	111	17	used	use	VERB
fcis-10369	111	18	for	for	ADP
fcis-10369	111	19	feature	feature	NOUN
fcis-10369	111	20	fusion	fusion	NOUN
fcis-10369	111	21	.	.	PUNCT
fcis-10369	112	1	the	the	DET
fcis-10369	112	2	concatenate	concatenate	NOUN
fcis-10369	112	3	function	function	NOUN
fcis-10369	112	4	concatenates	concatenate	VERB
fcis-10369	112	5	the	the	DET
fcis-10369	112	6	feature	feature	NOUN
fcis-10369	112	7	maps	map	NOUN
fcis-10369	112	8	of	of	ADP
fcis-10369	112	9	different	different	ADJ
fcis-10369	112	10	scales	scale	NOUN
fcis-10369	112	11	along	along	ADP
fcis-10369	112	12	the	the	DET
fcis-10369	112	13	channel	channel	NOUN
fcis-10369	112	14	dimension	dimension	NOUN
fcis-10369	112	15	,	,	PUNCT
fcis-10369	112	16	while	while	SCONJ
fcis-10369	112	17	the	the	DET
fcis-10369	112	18	add	add	NOUN
fcis-10369	112	19	function	function	NOUN
fcis-10369	112	20	performs	perform	VERB
fcis-10369	112	21	element	element	ADJ
fcis-10369	112	22	-	-	ADJ
fcis-10369	112	23	wise	wise	ADJ
fcis-10369	112	24	addition	addition	NOUN
fcis-10369	112	25	to	to	PART
fcis-10369	112	26	combine	combine	VERB
fcis-10369	112	27	the	the	DET
fcis-10369	112	28	feature	feature	NOUN
fcis-10369	112	29	maps	map	NOUN
fcis-10369	112	30	of	of	ADP
fcis-10369	112	31	multiple	multiple	ADJ
fcis-10369	112	32	scales	scale	NOUN
fcis-10369	112	33	,	,	PUNCT
fcis-10369	112	34	preserving	preserve	VERB
fcis-10369	112	35	the	the	DET
fcis-10369	112	36	information	information	NOUN
fcis-10369	112	37	of	of	ADP
fcis-10369	112	38	the	the	DET
fcis-10369	112	39	original	original	ADJ
fcis-10369	112	40	features	feature	NOUN
fcis-10369	112	41	.	.	PUNCT
fcis-10369	113	1	therefore	therefore	ADV
fcis-10369	113	2	,	,	PUNCT
fcis-10369	113	3	this	this	DET
fcis-10369	113	4	study	study	NOUN
fcis-10369	113	5	adopts	adopt	VERB
fcis-10369	113	6	the	the	DET
fcis-10369	113	7	add	add	VERB
fcis-10369	113	8	function	function	NOUN
fcis-10369	113	9	for	for	ADP
fcis-10369	113	10	the	the	DET
fcis-10369	113	11	fusion	fusion	NOUN
fcis-10369	113	12	of	of	ADP
fcis-10369	113	13	multi	multi	ADJ
fcis-10369	113	14	-	-	ADJ
fcis-10369	113	15	scale	scale	ADJ
fcis-10369	113	16	convolutional	convolutional	ADJ
fcis-10369	113	17	layers	layer	NOUN
fcis-10369	113	18	.	.	PUNCT
fcis-10369	114	1	it	it	PRON
fcis-10369	114	2	is	be	AUX
fcis-10369	114	3	important	important	ADJ
fcis-10369	114	4	to	to	PART
fcis-10369	114	5	note	note	VERB
fcis-10369	114	6	that	that	SCONJ
fcis-10369	114	7	when	when	SCONJ
fcis-10369	114	8	using	use	VERB
fcis-10369	114	9	the	the	DET
fcis-10369	114	10	add	add	ADJ
fcis-10369	114	11	function	function	NOUN
fcis-10369	114	12	to	to	PART
fcis-10369	114	13	merge	merge	VERB
fcis-10369	114	14	filters	filter	NOUN
fcis-10369	114	15	of	of	ADP
fcis-10369	114	16	different	different	ADJ
fcis-10369	114	17	scales	scale	NOUN
fcis-10369	114	18	,	,	PUNCT
fcis-10369	114	19	they	they	PRON
fcis-10369	114	20	need	need	VERB
fcis-10369	114	21	to	to	PART
fcis-10369	114	22	be	be	AUX
fcis-10369	114	23	adjusted	adjust	VERB
fcis-10369	114	24	to	to	PART
fcis-10369	114	25	have	have	VERB
fcis-10369	114	26	the	the	DET
fcis-10369	114	27	same	same	ADJ
fcis-10369	114	28	number	number	NOUN
fcis-10369	114	29	of	of	ADP
fcis-10369	114	30	channels	channel	NOUN
fcis-10369	114	31	.	.	PUNCT
fcis-10369	115	1	3.2.2	3.2.2	NUM
fcis-10369	115	2	.	.	PUNCT
fcis-10369	116	1	identity	identity	NOUN
fcis-10369	116	2	mapping	map	VERB
fcis-10369	116	3	the	the	DET
fcis-10369	116	4	identity	identity	NOUN
fcis-10369	116	5	mapping	mapping	NOUN
fcis-10369	116	6	,	,	PUNCT
fcis-10369	116	7	also	also	ADV
fcis-10369	116	8	known	know	VERB
fcis-10369	116	9	as	as	ADP
fcis-10369	116	10	skip	skip	ADJ
fcis-10369	116	11	connection	connection	NOUN
fcis-10369	116	12	,	,	PUNCT
fcis-10369	116	13	refers	refer	VERB
fcis-10369	116	14	to	to	ADP
fcis-10369	116	15	directly	directly	ADV
fcis-10369	116	16	passing	pass	VERB
fcis-10369	116	17	the	the	DET
fcis-10369	116	18	input	input	NOUN
fcis-10369	116	19	to	to	ADP
fcis-10369	116	20	the	the	DET
fcis-10369	116	21	output	output	NOUN
fcis-10369	116	22	of	of	ADP
fcis-10369	116	23	certain	certain	ADJ
fcis-10369	116	24	layers	layer	NOUN
fcis-10369	116	25	in	in	ADP
fcis-10369	116	26	a	a	DET
fcis-10369	116	27	neural	neural	ADJ
fcis-10369	116	28	network	network	NOUN
fcis-10369	116	29	.	.	PUNCT
fcis-10369	117	1	it	it	PRON
fcis-10369	117	2	introduces	introduce	VERB
fcis-10369	117	3	shortcut	shortcut	NOUN
fcis-10369	117	4	connections	connection	NOUN
fcis-10369	117	5	in	in	ADP
fcis-10369	117	6	deep	deep	ADJ
fcis-10369	117	7	neural	neural	ADJ
fcis-10369	117	8	networks	network	NOUN
fcis-10369	117	9	to	to	PART
fcis-10369	117	10	enhance	enhance	VERB
fcis-10369	117	11	model	model	NOUN
fcis-10369	117	12	convergence	convergence	NOUN
fcis-10369	117	13	and	and	CCONJ
fcis-10369	117	14	accelerate	accelerate	VERB
fcis-10369	117	15	training	training	NOUN
fcis-10369	117	16	.	.	PUNCT
fcis-10369	118	1	the	the	DET
fcis-10369	118	2	expression	expression	NOUN
fcis-10369	118	3	for	for	ADP
fcis-10369	118	4	the	the	DET
fcis-10369	118	5	identity	identity	NOUN
fcis-10369	118	6	mapping	mapping	NOUN
fcis-10369	118	7	is	be	AUX
fcis-10369	118	8	shown	show	VERB
fcis-10369	118	9	in	in	ADP
fcis-10369	118	10	equation	equation	NOUN
fcis-10369	118	11	(	(	PUNCT
fcis-10369	118	12	1	1	NUM
fcis-10369	118	13	)	)	PUNCT
fcis-10369	118	14	.	.	PUNCT
fcis-10369	119	1	(	(	PUNCT
fcis-10369	119	2	)	)	PUNCT
fcis-10369	119	3	y	y	PROPN
fcis-10369	119	4	f	f	PROPN
fcis-10369	119	5	x	x	X
fcis-10369	120	1	x	x	PROPN
fcis-10369	120	2			X
fcis-10369	120	3	(	(	PUNCT
fcis-10369	120	4	1	1	X
fcis-10369	120	5	)	)	PUNCT
fcis-10369	120	6	where	where	SCONJ
fcis-10369	120	7	x	x	PRON
fcis-10369	120	8	represents	represent	VERB
fcis-10369	120	9	the	the	DET
fcis-10369	120	10	input	input	NOUN
fcis-10369	120	11	and	and	CCONJ
fcis-10369	120	12	(	(	PUNCT
fcis-10369	120	13	)	)	PUNCT
fcis-10369	120	14	f	f	X
fcis-10369	120	15	x	x	PRON
fcis-10369	120	16	represents	represent	VERB
fcis-10369	120	17	the	the	DET
fcis-10369	120	18	transformation	transformation	NOUN
fcis-10369	120	19	operation	operation	NOUN
fcis-10369	120	20	of	of	ADP
fcis-10369	120	21	the	the	DET
fcis-10369	120	22	network	network	NOUN
fcis-10369	120	23	.	.	PUNCT
fcis-10369	121	1	through	through	ADP
fcis-10369	121	2	the	the	DET
fcis-10369	121	3	identity	identity	NOUN
fcis-10369	121	4	mapping	mapping	NOUN
fcis-10369	121	5	,	,	PUNCT
fcis-10369	121	6	the	the	DET
fcis-10369	121	7	network	network	NOUN
fcis-10369	121	8	adds	add	VERB
fcis-10369	121	9	the	the	DET
fcis-10369	121	10	input	input	NOUN
fcis-10369	121	11	to	to	ADP
fcis-10369	121	12	the	the	DET
fcis-10369	121	13	output	output	NOUN
fcis-10369	121	14	of	of	ADP
fcis-10369	121	15	the	the	DET
fcis-10369	121	16	transformation	transformation	NOUN
fcis-10369	121	17	operation	operation	NOUN
fcis-10369	121	18	to	to	PART
fcis-10369	121	19	obtain	obtain	VERB
fcis-10369	121	20	the	the	DET
fcis-10369	121	21	final	final	ADJ
fcis-10369	121	22	output	output	NOUN
fcis-10369	121	23	y	y	PROPN
fcis-10369	121	24	.	.	PUNCT
fcis-10369	122	1	even	even	ADV
fcis-10369	122	2	if	if	SCONJ
fcis-10369	122	3	the	the	DET
fcis-10369	122	4	transformation	transformation	NOUN
fcis-10369	122	5	operation	operation	NOUN
fcis-10369	122	6	of	of	ADP
fcis-10369	122	7	the	the	DET
fcis-10369	122	8	network	network	NOUN
fcis-10369	122	9	can	can	AUX
fcis-10369	122	10	not	not	PART
fcis-10369	122	11	perfectly	perfectly	ADV
fcis-10369	122	12	fit	fit	VERB
fcis-10369	122	13	the	the	DET
fcis-10369	122	14	data	datum	NOUN
fcis-10369	122	15	,	,	PUNCT
fcis-10369	122	16	the	the	DET
fcis-10369	122	17	information	information	NOUN
fcis-10369	122	18	from	from	ADP
fcis-10369	122	19	the	the	DET
fcis-10369	122	20	input	input	NOUN
fcis-10369	122	21	is	be	AUX
fcis-10369	122	22	still	still	ADV
fcis-10369	122	23	preserved	preserve	VERB
fcis-10369	122	24	and	and	CCONJ
fcis-10369	122	25	transmitted	transmit	VERB
fcis-10369	122	26	to	to	ADP
fcis-10369	122	27	the	the	DET
fcis-10369	122	28	output	output	NOUN
fcis-10369	122	29	through	through	ADP
fcis-10369	122	30	the	the	DET
fcis-10369	122	31	identity	identity	NOUN
fcis-10369	122	32	mapping	mapping	NOUN
fcis-10369	122	33	,	,	PUNCT
fcis-10369	122	34	retaining	retain	VERB
fcis-10369	122	35	the	the	DET
fcis-10369	122	36	original	original	ADJ
fcis-10369	122	37	input	input	NOUN
fcis-10369	122	38	information	information	NOUN
fcis-10369	122	39	.	.	PUNCT
fcis-10369	123	1	additionally	additionally	ADV
fcis-10369	123	2	,	,	PUNCT
fcis-10369	123	3	the	the	DET
fcis-10369	123	4	127	127	NUM
fcis-10369	123	5	identity	identity	NOUN
fcis-10369	123	6	mapping	mapping	NOUN
fcis-10369	123	7	does	do	AUX
fcis-10369	123	8	not	not	PART
fcis-10369	123	9	introduce	introduce	VERB
fcis-10369	123	10	additional	additional	ADJ
fcis-10369	123	11	parameters	parameter	NOUN
fcis-10369	123	12	or	or	CCONJ
fcis-10369	123	13	computational	computational	ADJ
fcis-10369	123	14	complexity	complexity	NOUN
fcis-10369	123	15	,	,	PUNCT
fcis-10369	123	16	yet	yet	CCONJ
fcis-10369	123	17	it	it	PRON
fcis-10369	123	18	helps	help	VERB
fcis-10369	123	19	the	the	DET
fcis-10369	123	20	network	network	NOUN
fcis-10369	123	21	learn	learn	VERB
fcis-10369	123	22	and	and	CCONJ
fcis-10369	123	23	optimize	optimize	VERB
fcis-10369	123	24	more	more	ADV
fcis-10369	123	25	effectively	effectively	ADV
fcis-10369	123	26	.	.	PUNCT
fcis-10369	124	1	by	by	ADP
fcis-10369	124	2	utilizing	utilize	VERB
fcis-10369	124	3	the	the	DET
fcis-10369	124	4	identity	identity	NOUN
fcis-10369	124	5	mapping	mapping	NOUN
fcis-10369	124	6	,	,	PUNCT
fcis-10369	124	7	the	the	DET
fcis-10369	124	8	network	network	NOUN
fcis-10369	124	9	can	can	AUX
fcis-10369	124	10	be	be	AUX
fcis-10369	124	11	stacked	stack	VERB
fcis-10369	124	12	deeper	deeply	ADV
fcis-10369	124	13	,	,	PUNCT
fcis-10369	124	14	improving	improve	VERB
fcis-10369	124	15	the	the	DET
fcis-10369	124	16	representation	representation	NOUN
fcis-10369	124	17	power	power	NOUN
fcis-10369	124	18	and	and	CCONJ
fcis-10369	124	19	performance	performance	NOUN
fcis-10369	124	20	of	of	ADP
fcis-10369	124	21	the	the	DET
fcis-10369	124	22	network	network	NOUN
fcis-10369	124	23	.	.	PUNCT
fcis-10369	125	1	4	4	X
fcis-10369	125	2	.	.	X
fcis-10369	125	3	experimental	experimental	ADJ
fcis-10369	125	4	setup	setup	NOUN
fcis-10369	125	5	and	and	CCONJ
fcis-10369	125	6	analysis	analysis	NOUN
fcis-10369	125	7	4.1	4.1	NUM
fcis-10369	125	8	.	.	PUNCT
fcis-10369	126	1	dataset	dataset	ADJ
fcis-10369	126	2	description	description	NOUN
fcis-10369	126	3	the	the	DET
fcis-10369	126	4	nsl	nsl	PROPN
fcis-10369	126	5	-	-	PUNCT
fcis-10369	126	6	kdd	kdd	PROPN
fcis-10369	126	7	dataset	dataset	NOUN
fcis-10369	126	8	is	be	AUX
fcis-10369	126	9	an	an	DET
fcis-10369	126	10	improvement	improvement	NOUN
fcis-10369	126	11	and	and	CCONJ
fcis-10369	126	12	enhancement	enhancement	NOUN
fcis-10369	126	13	of	of	ADP
fcis-10369	126	14	the	the	DET
fcis-10369	126	15	kddcup99	kddcup99	PROPN
fcis-10369	126	16	dataset	dataset	PROPN
fcis-10369	126	17	,	,	PUNCT
fcis-10369	126	18	aiming	aim	VERB
fcis-10369	126	19	to	to	PART
fcis-10369	126	20	improve	improve	VERB
fcis-10369	126	21	the	the	DET
fcis-10369	126	22	accuracy	accuracy	NOUN
fcis-10369	126	23	and	and	CCONJ
fcis-10369	126	24	reliability	reliability	NOUN
fcis-10369	126	25	of	of	ADP
fcis-10369	126	26	network	network	NOUN
fcis-10369	126	27	intrusion	intrusion	NOUN
fcis-10369	126	28	detection	detection	NOUN
fcis-10369	126	29	.	.	PUNCT
fcis-10369	127	1	the	the	DET
fcis-10369	127	2	dataset	dataset	NOUN
fcis-10369	127	3	contains	contain	VERB
fcis-10369	127	4	a	a	DET
fcis-10369	127	5	total	total	NOUN
fcis-10369	127	6	of	of	ADP
fcis-10369	127	7	41	41	NUM
fcis-10369	127	8	network	network	NOUN
fcis-10369	127	9	traffic	traffic	NOUN
fcis-10369	127	10	features	feature	NOUN
fcis-10369	127	11	,	,	PUNCT
fcis-10369	127	12	including	include	VERB
fcis-10369	127	13	34	34	NUM
fcis-10369	127	14	continuous	continuous	ADJ
fcis-10369	127	15	features	feature	NOUN
fcis-10369	127	16	,	,	PUNCT
fcis-10369	127	17	6	6	NUM
fcis-10369	127	18	discrete	discrete	ADJ
fcis-10369	127	19	features	feature	NOUN
fcis-10369	127	20	,	,	PUNCT
fcis-10369	127	21	and	and	CCONJ
fcis-10369	127	22	1	1	NUM
fcis-10369	127	23	binary	binary	ADJ
fcis-10369	127	24	feature	feature	NOUN
fcis-10369	127	25	.	.	PUNCT
fcis-10369	128	1	the	the	DET
fcis-10369	128	2	nsl	nsl	PROPN
fcis-10369	128	3	-	-	PUNCT
fcis-10369	128	4	kdd	kdd	PROPN
fcis-10369	128	5	dataset	dataset	NOUN
fcis-10369	128	6	is	be	AUX
fcis-10369	128	7	a	a	DET
fcis-10369	128	8	commonly	commonly	ADV
fcis-10369	128	9	used	use	VERB
fcis-10369	128	10	dataset	dataset	NOUN
fcis-10369	128	11	in	in	ADP
fcis-10369	128	12	the	the	DET
fcis-10369	128	13	field	field	NOUN
fcis-10369	128	14	of	of	ADP
fcis-10369	128	15	network	network	NOUN
fcis-10369	128	16	intrusion	intrusion	NOUN
fcis-10369	128	17	detection	detection	NOUN
fcis-10369	128	18	.	.	PUNCT
fcis-10369	129	1	in	in	ADP
fcis-10369	129	2	this	this	DET
fcis-10369	129	3	study	study	NOUN
fcis-10369	129	4	,	,	PUNCT
fcis-10369	129	5	the	the	DET
fcis-10369	129	6	kddtrain+	kddtrain+	X
fcis-10369	129	7	and	and	CCONJ
fcis-10369	129	8	kddtest+	kddtest+	NOUN
fcis-10369	129	9	subsets	subset	NOUN
fcis-10369	129	10	of	of	ADP
fcis-10369	129	11	the	the	DET
fcis-10369	129	12	nsl	nsl	PROPN
fcis-10369	129	13	-	-	PUNCT
fcis-10369	129	14	kdd	kdd	PROPN
fcis-10369	129	15	dataset	dataset	NOUN
fcis-10369	129	16	were	be	AUX
fcis-10369	129	17	used	use	VERB
fcis-10369	129	18	as	as	ADP
fcis-10369	129	19	the	the	DET
fcis-10369	129	20	training	training	NOUN
fcis-10369	129	21	and	and	CCONJ
fcis-10369	129	22	testing	testing	NOUN
fcis-10369	129	23	sets	set	NOUN
fcis-10369	129	24	for	for	ADP
fcis-10369	129	25	the	the	DET
fcis-10369	129	26	model	model	NOUN
fcis-10369	129	27	,	,	PUNCT
fcis-10369	129	28	respectively	respectively	ADV
fcis-10369	129	29	.	.	PUNCT
fcis-10369	130	1	the	the	DET
fcis-10369	130	2	class	class	NOUN
fcis-10369	130	3	distribution	distribution	NOUN
fcis-10369	130	4	and	and	CCONJ
fcis-10369	130	5	quantity	quantity	NOUN
fcis-10369	130	6	are	be	AUX
fcis-10369	130	7	shown	show	VERB
fcis-10369	130	8	in	in	ADP
fcis-10369	130	9	table	table	NOUN
fcis-10369	130	10	1	1	NUM
fcis-10369	130	11	.	.	PUNCT
fcis-10369	130	12	table	table	NOUN
fcis-10369	130	13	1	1	NUM
fcis-10369	130	14	.	.	PUNCT
fcis-10369	130	15	nsl	nsl	PROPN
fcis-10369	130	16	-	-	PUNCT
fcis-10369	130	17	kdd	kdd	PROPN
fcis-10369	130	18	categories	category	NOUN
fcis-10369	130	19	and	and	CCONJ
fcis-10369	130	20	quantities	quantity	NOUN
fcis-10369	130	21	total	total	VERB
fcis-10369	130	22	normal	normal	ADJ
fcis-10369	130	23	dos	do	NOUN
fcis-10369	130	24	probe	probe	VERB
fcis-10369	130	25	r2l	r2l	NOUN
fcis-10369	130	26	u2l	u2l	PROPN
fcis-10369	130	27	train	train	VERB
fcis-10369	130	28	125973	125973	NUM
fcis-10369	130	29	67343	67343	NUM
fcis-10369	130	30	45927	45927	NUM
fcis-10369	130	31	11656	11656	NUM
fcis-10369	130	32	995	995	NUM
fcis-10369	130	33	52	52	NUM
fcis-10369	130	34	test	test	NOUN
fcis-10369	130	35	22544	22544	NUM
fcis-10369	130	36	9711	9711	NUM
fcis-10369	130	37	7458	7458	NUM
fcis-10369	130	38	2421	2421	NUM
fcis-10369	130	39	2754	2754	NUM
fcis-10369	130	40	200	200	NUM
fcis-10369	130	41	the	the	DET
fcis-10369	130	42	unsw	unsw	PROPN
fcis-10369	130	43	-	-	PUNCT
fcis-10369	130	44	nb15	nb15	PROPN
fcis-10369	130	45	dataset	dataset	NOUN
fcis-10369	130	46	is	be	AUX
fcis-10369	130	47	a	a	DET
fcis-10369	130	48	network	network	NOUN
fcis-10369	130	49	intrusion	intrusion	NOUN
fcis-10369	130	50	detection	detection	NOUN
fcis-10369	130	51	dataset	dataset	VERB
fcis-10369	130	52	developed	develop	VERB
fcis-10369	130	53	by	by	ADP
fcis-10369	130	54	the	the	DET
fcis-10369	130	55	university	university	NOUN
fcis-10369	130	56	of	of	ADP
fcis-10369	130	57	new	new	ADJ
fcis-10369	130	58	south	south	PROPN
fcis-10369	130	59	wales	wales	PROPN
fcis-10369	130	60	in	in	ADP
fcis-10369	130	61	australia	australia	PROPN
fcis-10369	130	62	.	.	PUNCT
fcis-10369	131	1	it	it	PRON
fcis-10369	131	2	consists	consist	VERB
fcis-10369	131	3	of	of	ADP
fcis-10369	131	4	49	49	NUM
fcis-10369	131	5	network	network	NOUN
fcis-10369	131	6	traffic	traffic	NOUN
fcis-10369	131	7	features	feature	NOUN
fcis-10369	131	8	,	,	PUNCT
fcis-10369	131	9	including	include	VERB
fcis-10369	131	10	45	45	NUM
fcis-10369	131	11	continuous	continuous	ADJ
fcis-10369	131	12	features	feature	NOUN
fcis-10369	131	13	,	,	PUNCT
fcis-10369	131	14	3	3	NUM
fcis-10369	131	15	discrete	discrete	ADJ
fcis-10369	131	16	features	feature	NOUN
fcis-10369	131	17	,	,	PUNCT
fcis-10369	131	18	and	and	CCONJ
fcis-10369	131	19	1	1	NUM
fcis-10369	131	20	binary	binary	ADJ
fcis-10369	131	21	feature	feature	NOUN
fcis-10369	131	22	.	.	PUNCT
fcis-10369	132	1	unsw	unsw	NOUN
fcis-10369	132	2	-	-	PUNCT
fcis-10369	132	3	nb15	nb15	PROPN
fcis-10369	132	4	contains	contain	VERB
fcis-10369	132	5	a	a	DET
fcis-10369	132	6	wider	wide	ADJ
fcis-10369	132	7	range	range	NOUN
fcis-10369	132	8	of	of	ADP
fcis-10369	132	9	attack	attack	NOUN
fcis-10369	132	10	types	type	NOUN
fcis-10369	132	11	and	and	CCONJ
fcis-10369	132	12	network	network	NOUN
fcis-10369	132	13	traffic	traffic	NOUN
fcis-10369	132	14	features	feature	NOUN
fcis-10369	132	15	,	,	PUNCT
fcis-10369	132	16	allowing	allow	VERB
fcis-10369	132	17	for	for	ADP
fcis-10369	132	18	a	a	DET
fcis-10369	132	19	more	more	ADV
fcis-10369	132	20	comprehensive	comprehensive	ADJ
fcis-10369	132	21	evaluation	evaluation	NOUN
fcis-10369	132	22	of	of	ADP
fcis-10369	132	23	intrusion	intrusion	NOUN
fcis-10369	132	24	detection	detection	NOUN
fcis-10369	132	25	models	model	NOUN
fcis-10369	132	26	.	.	PUNCT
fcis-10369	133	1	in	in	ADP
fcis-10369	133	2	the	the	DET
fcis-10369	133	3	experiments	experiment	NOUN
fcis-10369	133	4	,	,	PUNCT
fcis-10369	133	5	the	the	DET
fcis-10369	133	6	unsw_nb15_training	unsw_nb15_traine	VERB
fcis-10369	133	7	-	-	PUNCT
fcis-10369	133	8	set	set	ADJ
fcis-10369	133	9	and	and	CCONJ
fcis-10369	133	10	unsw_nb15_testing	unsw_nb15_testing	NOUN
fcis-10369	133	11	-	-	PUNCT
fcis-10369	133	12	set	set	NOUN
fcis-10369	133	13	provided	provide	VERB
fcis-10369	133	14	by	by	ADP
fcis-10369	133	15	the	the	DET
fcis-10369	133	16	dataset	dataset	ADJ
fcis-10369	133	17	creators	creator	NOUN
fcis-10369	133	18	were	be	AUX
fcis-10369	133	19	used	use	VERB
fcis-10369	133	20	as	as	ADP
fcis-10369	133	21	the	the	DET
fcis-10369	133	22	training	training	NOUN
fcis-10369	133	23	and	and	CCONJ
fcis-10369	133	24	testing	testing	NOUN
fcis-10369	133	25	sets	set	NOUN
fcis-10369	133	26	,	,	PUNCT
fcis-10369	133	27	respectively	respectively	ADV
fcis-10369	133	28	.	.	PUNCT
fcis-10369	134	1	the	the	DET
fcis-10369	134	2	class	class	NOUN
fcis-10369	134	3	distribution	distribution	NOUN
fcis-10369	134	4	and	and	CCONJ
fcis-10369	134	5	quantity	quantity	NOUN
fcis-10369	134	6	are	be	AUX
fcis-10369	134	7	shown	show	VERB
fcis-10369	134	8	in	in	ADP
fcis-10369	134	9	table	table	NOUN
fcis-10369	134	10	2	2	NUM
fcis-10369	134	11	.	.	PUNCT
fcis-10369	134	12	table	table	NOUN
fcis-10369	134	13	2	2	NUM
fcis-10369	134	14	.	.	PUNCT
fcis-10369	134	15	unsw	unsw	PROPN
fcis-10369	134	16	-	-	PUNCT
fcis-10369	134	17	nb15	nb15	PROPN
fcis-10369	134	18	categories	category	NOUN
fcis-10369	134	19	and	and	CCONJ
fcis-10369	134	20	quantities	quantity	NOUN
fcis-10369	134	21	category	category	NOUN
fcis-10369	134	22	training	train	VERB
fcis-10369	134	23	set	set	NOUN
fcis-10369	134	24	testing	testing	NOUN
fcis-10369	134	25	set	set	VERB
fcis-10369	134	26	normal	normal	ADJ
fcis-10369	134	27	56000	56000	NUM
fcis-10369	134	28	37000	37000	NUM
fcis-10369	134	29	analysis	analysis	NOUN
fcis-10369	134	30	2000	2000	NUM
fcis-10369	134	31	677	677	NUM
fcis-10369	134	32	backdoor	backdoor	NOUN
fcis-10369	134	33	1746	1746	NUM
fcis-10369	134	34	583	583	NUM
fcis-10369	134	35	dos	do	NOUN
fcis-10369	134	36	12264	12264	NUM
fcis-10369	134	37	4089	4089	NUM
fcis-10369	134	38	exploits	exploit	VERB
fcis-10369	134	39	33393	33393	NUM
fcis-10369	134	40	11132	11132	NUM
fcis-10369	134	41	fuzzers	fuzzer	NOUN
fcis-10369	134	42	18184	18184	NUM
fcis-10369	134	43	6062	6062	NUM
fcis-10369	134	44	generic	generic	ADJ
fcis-10369	134	45	40000	40000	NUM
fcis-10369	134	46	18871	18871	NUM
fcis-10369	134	47	reconnaissance	reconnaissance	NOUN
fcis-10369	134	48	10491	10491	NUM
fcis-10369	134	49	3496	3496	NUM
fcis-10369	134	50	shellcode	shellcode	NOUN
fcis-10369	134	51	1133	1133	NUM
fcis-10369	134	52	378	378	NUM
fcis-10369	134	53	worms	worm	NOUN
fcis-10369	134	54	130	130	NUM
fcis-10369	134	55	44	44	NUM
fcis-10369	134	56	total	total	NOUN
fcis-10369	134	57	175341	175341	NUM
fcis-10369	134	58	82332	82332	NUM
fcis-10369	134	59	4.2	4.2	NUM
fcis-10369	134	60	.	.	PUNCT
fcis-10369	135	1	preprocessing	preprocesse	VERB
fcis-10369	135	2	data	datum	NOUN
fcis-10369	135	3	preprocessing	preprocessing	NOUN
fcis-10369	135	4	involves	involve	NOUN
fcis-10369	135	5	performing	perform	VERB
fcis-10369	135	6	numerical	numerical	ADJ
fcis-10369	135	7	encoding	encoding	NOUN
fcis-10369	135	8	and	and	CCONJ
fcis-10369	135	9	normalization	normalization	NOUN
fcis-10369	135	10	on	on	ADP
fcis-10369	135	11	two	two	NUM
fcis-10369	135	12	benchmark	benchmark	ADJ
fcis-10369	135	13	datasets	dataset	NOUN
fcis-10369	135	14	,	,	PUNCT
fcis-10369	135	15	nsl	nsl	PROPN
fcis-10369	135	16	-	-	PUNCT
fcis-10369	135	17	kdd	kdd	PROPN
fcis-10369	135	18	and	and	CCONJ
fcis-10369	135	19	unsw	unsw	PROPN
fcis-10369	135	20	-	-	PUNCT
fcis-10369	135	21	nb15	nb15	PROPN
fcis-10369	135	22	.	.	PUNCT
fcis-10369	136	1	the	the	DET
fcis-10369	136	2	specific	specific	ADJ
fcis-10369	136	3	steps	step	NOUN
fcis-10369	136	4	are	be	AUX
fcis-10369	136	5	as	as	SCONJ
fcis-10369	136	6	follows	follow	VERB
fcis-10369	136	7	:	:	PUNCT
fcis-10369	136	8	(	(	PUNCT
fcis-10369	136	9	1	1	X
fcis-10369	136	10	)	)	PUNCT
fcis-10369	136	11	numerical	numerical	PROPN
fcis-10369	136	12	encoding	encoding	NOUN
fcis-10369	136	13	.	.	PUNCT
fcis-10369	137	1	non	non	ADJ
fcis-10369	137	2	-	-	ADJ
fcis-10369	137	3	numerical	numerical	ADJ
fcis-10369	137	4	features	feature	NOUN
fcis-10369	137	5	in	in	ADP
fcis-10369	137	6	the	the	DET
fcis-10369	137	7	dataset	dataset	NOUN
fcis-10369	137	8	need	need	NOUN
fcis-10369	137	9	to	to	PART
fcis-10369	137	10	be	be	AUX
fcis-10369	137	11	converted	convert	VERB
fcis-10369	137	12	into	into	ADP
fcis-10369	137	13	numerical	numerical	ADJ
fcis-10369	137	14	features	feature	NOUN
fcis-10369	137	15	.	.	PUNCT
fcis-10369	138	1	taking	take	VERB
fcis-10369	138	2	the	the	DET
fcis-10369	138	3	unsw	unsw	NOUN
fcis-10369	138	4	-	-	PUNCT
fcis-10369	138	5	nb15	nb15	PROPN
fcis-10369	138	6	dataset	dataset	VERB
fcis-10369	138	7	as	as	ADP
fcis-10369	138	8	an	an	DET
fcis-10369	138	9	example	example	NOUN
fcis-10369	138	10	,	,	PUNCT
fcis-10369	138	11	'	'	PUNCT
fcis-10369	138	12	proto	proto	NOUN
fcis-10369	138	13	'	'	PUNCT
fcis-10369	138	14	,	,	PUNCT
fcis-10369	138	15	'	'	PUNCT
fcis-10369	138	16	state	state	NOUN
fcis-10369	138	17	'	'	PUNCT
fcis-10369	138	18	,	,	PUNCT
fcis-10369	138	19	and	and	CCONJ
fcis-10369	138	20	'	'	PUNCT
fcis-10369	138	21	service	service	NOUN
fcis-10369	138	22	'	'	PUNCT
fcis-10369	138	23	are	be	AUX
fcis-10369	138	24	non	non	ADJ
fcis-10369	138	25	-	-	ADJ
fcis-10369	138	26	numerical	numerical	ADJ
fcis-10369	138	27	features	feature	NOUN
fcis-10369	138	28	,	,	PUNCT
fcis-10369	138	29	while	while	SCONJ
fcis-10369	138	30	the	the	DET
fcis-10369	138	31	rest	rest	NOUN
fcis-10369	138	32	are	be	AUX
fcis-10369	138	33	numerical	numerical	ADJ
fcis-10369	138	34	features	feature	NOUN
fcis-10369	138	35	.	.	PUNCT
fcis-10369	139	1	one	one	NUM
fcis-10369	139	2	way	way	NOUN
fcis-10369	139	3	to	to	PART
fcis-10369	139	4	handle	handle	VERB
fcis-10369	139	5	this	this	PRON
fcis-10369	139	6	is	be	AUX
fcis-10369	139	7	by	by	ADP
fcis-10369	139	8	using	use	VERB
fcis-10369	139	9	onehot	onehot	ADJ
fcis-10369	139	10	encoding	encoding	NOUN
fcis-10369	139	11	.	.	PUNCT
fcis-10369	140	1	'	'	PUNCT
fcis-10369	140	2	proto	proto	NOUN
fcis-10369	140	3	'	'	PUNCT
fcis-10369	140	4	represents	represent	VERB
fcis-10369	140	5	transaction	transaction	NOUN
fcis-10369	140	6	protocol	protocol	NOUN
fcis-10369	140	7	types	type	NOUN
fcis-10369	140	8	,	,	PUNCT
fcis-10369	140	9	including	include	VERB
fcis-10369	140	10	133	133	NUM
fcis-10369	140	11	different	different	ADJ
fcis-10369	140	12	types	type	NOUN
fcis-10369	140	13	such	such	ADJ
fcis-10369	140	14	as	as	ADP
fcis-10369	140	15	tcp	tcp	PROPN
fcis-10369	140	16	,	,	PUNCT
fcis-10369	140	17	udp	udp	NOUN
fcis-10369	140	18	,	,	PUNCT
fcis-10369	140	19	arp	arp	ADJ
fcis-10369	140	20	.	.	PUNCT
fcis-10369	141	1	'	'	PUNCT
fcis-10369	141	2	state	state	NOUN
fcis-10369	141	3	'	'	PUNCT
fcis-10369	141	4	represents	represent	VERB
fcis-10369	141	5	states	state	NOUN
fcis-10369	141	6	and	and	CCONJ
fcis-10369	141	7	related	related	ADJ
fcis-10369	141	8	protocols	protocol	NOUN
fcis-10369	141	9	,	,	PUNCT
fcis-10369	141	10	including	include	VERB
fcis-10369	141	11	13	13	NUM
fcis-10369	141	12	different	different	ADJ
fcis-10369	141	13	types	type	NOUN
fcis-10369	141	14	such	such	ADJ
fcis-10369	141	15	as	as	ADP
fcis-10369	141	16	con	con	PROPN
fcis-10369	141	17	,	,	PUNCT
fcis-10369	141	18	eco	eco	PROPN
fcis-10369	141	19	,	,	PUNCT
fcis-10369	141	20	rst	rst	PROPN
fcis-10369	141	21	.	.	PUNCT
fcis-10369	142	1	'	'	PUNCT
fcis-10369	142	2	service	service	NOUN
fcis-10369	142	3	'	'	PUNCT
fcis-10369	142	4	includes	include	VERB
fcis-10369	142	5	11	11	NUM
fcis-10369	142	6	different	different	ADJ
fcis-10369	142	7	types	type	NOUN
fcis-10369	142	8	of	of	ADP
fcis-10369	142	9	services	service	NOUN
fcis-10369	142	10	such	such	ADJ
fcis-10369	142	11	as	as	ADP
fcis-10369	142	12	http	http	ADJ
fcis-10369	142	13	,	,	PUNCT
fcis-10369	142	14	ftp	ftp	PROPN
fcis-10369	142	15	,	,	PUNCT
fcis-10369	142	16	ssh	ssh	PROPN
fcis-10369	142	17	,	,	PUNCT
fcis-10369	142	18	dns	dns	PROPN
fcis-10369	142	19	.	.	PUNCT
fcis-10369	143	1	(	(	PUNCT
fcis-10369	143	2	2	2	NUM
fcis-10369	143	3	)	)	PUNCT
fcis-10369	143	4	normalization	normalization	NOUN
fcis-10369	143	5	.	.	PUNCT
fcis-10369	144	1	due	due	ADP
fcis-10369	144	2	to	to	ADP
fcis-10369	144	3	the	the	DET
fcis-10369	144	4	significant	significant	ADJ
fcis-10369	144	5	differences	difference	NOUN
fcis-10369	144	6	in	in	ADP
fcis-10369	144	7	the	the	DET
fcis-10369	144	8	numerical	numerical	ADJ
fcis-10369	144	9	feature	feature	NOUN
fcis-10369	144	10	values	value	NOUN
fcis-10369	144	11	across	across	ADP
fcis-10369	144	12	dimensions	dimension	NOUN
fcis-10369	144	13	,	,	PUNCT
fcis-10369	144	14	for	for	ADP
fcis-10369	144	15	example	example	NOUN
fcis-10369	144	16	,	,	PUNCT
fcis-10369	144	17	the	the	DET
fcis-10369	144	18	feature	feature	NOUN
fcis-10369	144	19	'	'	PUNCT
fcis-10369	144	20	sybets	sybet	NOUN
fcis-10369	144	21	'	'	PART
fcis-10369	144	22	ranges	range	NOUN
fcis-10369	144	23	from	from	ADP
fcis-10369	144	24	0	0	NUM
fcis-10369	144	25	to	to	ADP
fcis-10369	144	26	129652	129652	NUM
fcis-10369	144	27	.	.	PUNCT
fcis-10369	145	1	to	to	PART
fcis-10369	145	2	scale	scale	VERB
fcis-10369	145	3	the	the	DET
fcis-10369	145	4	data	datum	NOUN
fcis-10369	145	5	to	to	ADP
fcis-10369	145	6	the	the	DET
fcis-10369	145	7	same	same	ADJ
fcis-10369	145	8	range	range	NOUN
fcis-10369	145	9	and	and	CCONJ
fcis-10369	145	10	ensure	ensure	VERB
fcis-10369	145	11	that	that	SCONJ
fcis-10369	145	12	different	different	ADJ
fcis-10369	145	13	features	feature	NOUN
fcis-10369	145	14	have	have	VERB
fcis-10369	145	15	equal	equal	ADJ
fcis-10369	145	16	impact	impact	NOUN
fcis-10369	145	17	on	on	ADP
fcis-10369	145	18	the	the	DET
fcis-10369	145	19	model	model	NOUN
fcis-10369	145	20	training	training	NOUN
fcis-10369	145	21	results	result	NOUN
fcis-10369	145	22	,	,	PUNCT
fcis-10369	145	23	min	min	ADJ
fcis-10369	145	24	-	-	ADJ
fcis-10369	145	25	max	max	PROPN
fcis-10369	145	26	normalization	normalization	NOUN
fcis-10369	145	27	is	be	AUX
fcis-10369	145	28	applied	apply	VERB
fcis-10369	145	29	after	after	ADP
fcis-10369	145	30	numerical	numerical	PROPN
fcis-10369	145	31	encoding	encoding	PROPN
fcis-10369	145	32	.	.	PUNCT
fcis-10369	146	1	the	the	DET
fcis-10369	146	2	normalization	normalization	NOUN
fcis-10369	146	3	operation	operation	NOUN
fcis-10369	146	4	is	be	AUX
fcis-10369	146	5	represented	represent	VERB
fcis-10369	146	6	by	by	ADP
fcis-10369	146	7	equation	equation	NOUN
fcis-10369	146	8	(	(	PUNCT
fcis-10369	146	9	2	2	NUM
fcis-10369	146	10	)	)	PUNCT
fcis-10369	146	11	.	.	PUNCT
fcis-10369	147	1	[	[	PUNCT
fcis-10369	147	2	]	]	X
fcis-10369	147	3	min	min	NOUN
fcis-10369	147	4	[	[	PUNCT
fcis-10369	147	5	]	]	X
fcis-10369	147	6	max	max	PROPN
fcis-10369	147	7	min	min	PROPN
fcis-10369	148	1	i	i	INTJ
fcis-10369	148	2	i	i	VERB
fcis-10369	148	3	x	x	VERB
fcis-10369	148	4	x	x	PUNCT
fcis-10369	148	5	x	x	PUNCT
fcis-10369	148	6	x	x	SYM
fcis-10369	148	7	x	x	X
fcis-10369	148	8			PROPN
fcis-10369	148	9			NUM
fcis-10369	148	10			NOUN
fcis-10369	148	11	(	(	PUNCT
fcis-10369	148	12	2	2	NUM
fcis-10369	148	13	)	)	PUNCT
fcis-10369	148	14	where	where	SCONJ
fcis-10369	148	15	[	[	PUNCT
fcis-10369	148	16	]	]	X
fcis-10369	148	17	ix	ix	VERB
fcis-10369	148	18	represents	represent	VERB
fcis-10369	148	19	the	the	DET
fcis-10369	148	20	feature	feature	NOUN
fcis-10369	148	21	value	value	NOUN
fcis-10369	148	22	to	to	PART
fcis-10369	148	23	be	be	AUX
fcis-10369	148	24	normalized	normalize	VERB
fcis-10369	148	25	,	,	PUNCT
fcis-10369	148	26	minx	minx	PROPN
fcis-10369	148	27	and	and	CCONJ
fcis-10369	148	28	maxx	maxx	PROPN
fcis-10369	148	29	represent	represent	VERB
fcis-10369	148	30	the	the	DET
fcis-10369	148	31	minimum	minimum	ADJ
fcis-10369	148	32	and	and	CCONJ
fcis-10369	148	33	maximum	maximum	ADJ
fcis-10369	148	34	values	value	NOUN
fcis-10369	148	35	of	of	ADP
fcis-10369	148	36	that	that	DET
fcis-10369	148	37	feature	feature	NOUN
fcis-10369	148	38	attribute	attribute	NOUN
fcis-10369	148	39	,	,	PUNCT
fcis-10369	148	40	respectively	respectively	ADV
fcis-10369	148	41	.	.	PUNCT
fcis-10369	149	1	4.3	4.3	NUM
fcis-10369	149	2	.	.	PUNCT
fcis-10369	150	1	experimental	experimental	ADJ
fcis-10369	150	2	settings	setting	NOUN
fcis-10369	150	3	table	table	VERB
fcis-10369	150	4	3	3	NUM
fcis-10369	150	5	.	.	PUNCT
fcis-10369	151	1	parameter	parameter	NOUN
fcis-10369	151	2	settings	setting	NOUN
fcis-10369	151	3	network	network	NOUN
fcis-10369	151	4	architecture	architecture	NOUN
fcis-10369	151	5	layers	layer	NOUN
fcis-10369	151	6	nsl	nsl	PROPN
fcis-10369	151	7	-	-	PUNCT
fcis-10369	151	8	kdd	kdd	PROPN
fcis-10369	151	9	unsw	unsw	PROPN
fcis-10369	151	10	-	-	PUNCT
fcis-10369	151	11	nb15	nb15	PROPN
fcis-10369	151	12	parameter	parameter	NOUN
fcis-10369	151	13	values	value	NOUN
fcis-10369	151	14	parameter	parameter	PROPN
fcis-10369	151	15	values	value	NOUN
fcis-10369	151	16	inputlayer	inputlayer	NOUN
fcis-10369	151	17	122,1	122,1	NUM
fcis-10369	151	18	196,1	196,1	NUM
fcis-10369	151	19	convolution1d	convolution1d	NOUN
fcis-10369	151	20	16,122	16,122	NUM
fcis-10369	151	21	,	,	PUNCT
fcis-10369	151	22	same	same	ADJ
fcis-10369	151	23	,	,	PUNCT
fcis-10369	151	24	relu	relu	NOUN
fcis-10369	151	25	16,196	16,196	NUM
fcis-10369	151	26	,	,	PUNCT
fcis-10369	151	27	same	same	ADJ
fcis-10369	151	28	,	,	PUNCT
fcis-10369	151	29	relu	relu	NOUN
fcis-10369	151	30	32,122	32,122	NUM
fcis-10369	151	31	,	,	PUNCT
fcis-10369	151	32	same	same	ADJ
fcis-10369	151	33	,	,	PUNCT
fcis-10369	151	34	relu	relu	NOUN
fcis-10369	151	35	16,196	16,196	NUM
fcis-10369	151	36	,	,	PUNCT
fcis-10369	151	37	same	same	ADJ
fcis-10369	151	38	,	,	PUNCT
fcis-10369	151	39	relu	relu	NOUN
fcis-10369	151	40	64,122	64,122	NUM
fcis-10369	151	41	,	,	PUNCT
fcis-10369	151	42	same	same	ADJ
fcis-10369	151	43	,	,	PUNCT
fcis-10369	151	44	relu	relu	NOUN
fcis-10369	151	45	16,196	16,196	NUM
fcis-10369	151	46	,	,	PUNCT
fcis-10369	151	47	same	same	ADJ
fcis-10369	151	48	,	,	PUNCT
fcis-10369	151	49	relu	relu	NOUN
fcis-10369	151	50	maxpooling	maxpoole	VERB
fcis-10369	151	51	5	5	NUM
fcis-10369	151	52	10	10	NUM
fcis-10369	151	53	batchnormalization	batchnormalization	NOUN
fcis-10369	151	54	bidirectional	bidirectional	NOUN
fcis-10369	151	55	(	(	PUNCT
fcis-10369	151	56	lstm	lstm	NOUN
fcis-10369	151	57	)	)	PUNCT
fcis-10369	151	58	61	61	NUM
fcis-10369	151	59	98	98	NUM
fcis-10369	151	60	maxpooling	maxpoole	VERB
fcis-10369	151	61	5	5	NUM
fcis-10369	151	62	5	5	NUM
fcis-10369	151	63	batchnormalization	batchnormalization	NOUN
fcis-10369	151	64	bidirectional	bidirectional	NOUN
fcis-10369	151	65	(	(	PUNCT
fcis-10369	151	66	lstm	lstm	NOUN
fcis-10369	151	67	)	)	PUNCT
fcis-10369	151	68	122	122	NUM
fcis-10369	151	69	128	128	NUM
fcis-10369	151	70	dropout	dropout	NOUN
fcis-10369	151	71	0.5	0.5	NUM
fcis-10369	151	72	0.2	0.2	NUM
fcis-10369	151	73	dense	dense	ADJ
fcis-10369	151	74	5	5	NUM
fcis-10369	151	75	10	10	NUM
fcis-10369	151	76	activation_layer	activation_layer	NOUN
fcis-10369	151	77	softmax	softmax	NOUN
fcis-10369	151	78	softmax	softmax	X
fcis-10369	151	79	128	128	NUM
fcis-10369	151	80	experimental	experimental	ADJ
fcis-10369	151	81	setup	setup	NOUN
fcis-10369	151	82	:	:	PUNCT
fcis-10369	151	83	hardware	hardware	NOUN
fcis-10369	151	84	environment	environment	NOUN
fcis-10369	151	85	:	:	PUNCT
fcis-10369	151	86	intel(r	intel(r	PROPN
fcis-10369	151	87	)	)	PUNCT
fcis-10369	151	88	core	core	NOUN
fcis-10369	151	89	i7	i7	NOUN
fcis-10369	151	90	-	-	PUNCT
fcis-10369	151	91	8550u	8550u	NUM
fcis-10369	151	92	@	@	ADP
fcis-10369	151	93	1.80ghz	1.80ghz	NUM
fcis-10369	151	94	quad	quad	ADV
fcis-10369	151	95	-	-	PUNCT
fcis-10369	151	96	core	core	NOUN
fcis-10369	151	97	,	,	PUNCT
fcis-10369	151	98	16	16	NUM
fcis-10369	151	99	gb	gb	NOUN
fcis-10369	151	100	ram	ram	NOUN
fcis-10369	151	101	.	.	PUNCT
fcis-10369	152	1	software	software	NOUN
fcis-10369	152	2	environment	environment	PROPN
fcis-10369	152	3	:	:	PUNCT
fcis-10369	152	4	windows	window	VERB
fcis-10369	152	5	10	10	NUM
fcis-10369	152	6	operating	operating	NOUN
fcis-10369	152	7	system	system	NOUN
fcis-10369	152	8	,	,	PUNCT
fcis-10369	152	9	tensorflow	tensorflow	NOUN
fcis-10369	152	10	2.1	2.1	NUM
fcis-10369	152	11	,	,	PUNCT
fcis-10369	152	12	and	and	CCONJ
fcis-10369	152	13	python	python	NOUN
fcis-10369	152	14	3.6.3	3.6.3	NUM
fcis-10369	152	15	.	.	PUNCT
fcis-10369	153	1	as	as	SCONJ
fcis-10369	153	2	described	describe	VERB
fcis-10369	153	3	in	in	ADP
fcis-10369	153	4	section	section	NOUN
fcis-10369	153	5	4.2	4.2	NUM
fcis-10369	153	6	,	,	PUNCT
fcis-10369	153	7	after	after	ADP
fcis-10369	153	8	data	data	NOUN
fcis-10369	153	9	preprocessing	preprocessing	NOUN
fcis-10369	153	10	,	,	PUNCT
fcis-10369	153	11	the	the	DET
fcis-10369	153	12	nsl	nsl	PROPN
fcis-10369	153	13	-	-	PUNCT
fcis-10369	153	14	kdd	kdd	PROPN
fcis-10369	153	15	dataset	dataset	NOUN
fcis-10369	153	16	has	have	VERB
fcis-10369	153	17	122	122	NUM
fcis-10369	153	18	dimensions	dimension	NOUN
fcis-10369	153	19	,	,	PUNCT
fcis-10369	153	20	and	and	CCONJ
fcis-10369	153	21	the	the	DET
fcis-10369	153	22	unsw	unsw	PROPN
fcis-10369	153	23	-	-	PUNCT
fcis-10369	153	24	nb15	nb15	PROPN
fcis-10369	153	25	dataset	dataset	NOUN
fcis-10369	153	26	has	have	VERB
fcis-10369	153	27	196	196	NUM
fcis-10369	153	28	dimensions	dimension	NOUN
fcis-10369	153	29	.	.	PUNCT
fcis-10369	154	1	due	due	ADP
fcis-10369	154	2	to	to	ADP
fcis-10369	154	3	the	the	DET
fcis-10369	154	4	need	need	NOUN
fcis-10369	154	5	for	for	ADP
fcis-10369	154	6	feature	feature	NOUN
fcis-10369	154	7	fusion	fusion	NOUN
fcis-10369	154	8	and	and	CCONJ
fcis-10369	154	9	the	the	DET
fcis-10369	154	10	introduction	introduction	NOUN
fcis-10369	154	11	of	of	ADP
fcis-10369	154	12	residual	residual	ADJ
fcis-10369	154	13	connections	connection	NOUN
fcis-10369	154	14	in	in	ADP
fcis-10369	154	15	the	the	DET
fcis-10369	154	16	msff	msff	NOUN
fcis-10369	154	17	model	model	NOUN
fcis-10369	154	18	,	,	PUNCT
fcis-10369	154	19	a	a	DET
fcis-10369	154	20	simple	simple	ADJ
fcis-10369	154	21	sequential	sequential	ADJ
fcis-10369	154	22	model	model	NOUN
fcis-10369	154	23	structure	structure	NOUN
fcis-10369	154	24	is	be	AUX
fcis-10369	154	25	not	not	PART
fcis-10369	154	26	sufficient	sufficient	ADJ
fcis-10369	154	27	.	.	PUNCT
fcis-10369	155	1	therefore	therefore	ADV
fcis-10369	155	2	,	,	PUNCT
fcis-10369	155	3	a	a	DET
fcis-10369	155	4	functional	functional	ADJ
fcis-10369	155	5	model	model	NOUN
fcis-10369	155	6	structure	structure	NOUN
fcis-10369	155	7	is	be	AUX
fcis-10369	155	8	adopted	adopt	VERB
fcis-10369	155	9	.	.	PUNCT
fcis-10369	156	1	the	the	DET
fcis-10369	156	2	msff	msff	NOUN
fcis-10369	156	3	model	model	NOUN
fcis-10369	156	4	consists	consist	VERB
fcis-10369	156	5	of	of	ADP
fcis-10369	156	6	multiple	multiple	ADJ
fcis-10369	156	7	one	one	NUM
fcis-10369	156	8	-	-	PUNCT
fcis-10369	156	9	dimensional	dimensional	ADJ
fcis-10369	156	10	convolutional	convolutional	ADJ
fcis-10369	156	11	layers	layer	NOUN
fcis-10369	156	12	with	with	ADP
fcis-10369	156	13	different	different	ADJ
fcis-10369	156	14	scales	scale	NOUN
fcis-10369	156	15	of	of	ADP
fcis-10369	156	16	filters	filter	NOUN
fcis-10369	156	17	(	(	PUNCT
fcis-10369	156	18	16	16	NUM
fcis-10369	156	19	,	,	PUNCT
fcis-10369	156	20	32	32	NUM
fcis-10369	156	21	,	,	PUNCT
fcis-10369	156	22	and	and	CCONJ
fcis-10369	156	23	64	64	NUM
fcis-10369	156	24	)	)	PUNCT
fcis-10369	156	25	and	and	CCONJ
fcis-10369	156	26	two	two	NUM
fcis-10369	156	27	bilstm	bilstm	NOUN
fcis-10369	156	28	layers	layer	NOUN
fcis-10369	156	29	.	.	PUNCT
fcis-10369	157	1	the	the	DET
fcis-10369	157	2	model	model	NOUN
fcis-10369	157	3	also	also	ADV
fcis-10369	157	4	incorporates	incorporate	VERB
fcis-10369	157	5	skip	skip	ADJ
fcis-10369	157	6	connections	connection	NOUN
fcis-10369	157	7	(	(	PUNCT
fcis-10369	157	8	identity	identity	NOUN
fcis-10369	157	9	mapping	mapping	NOUN
fcis-10369	157	10	)	)	PUNCT
fcis-10369	157	11	between	between	ADP
fcis-10369	157	12	the	the	DET
fcis-10369	157	13	input	input	NOUN
fcis-10369	157	14	and	and	CCONJ
fcis-10369	157	15	bilstm	bilstm	NOUN
fcis-10369	157	16	layers	layer	NOUN
fcis-10369	157	17	.	.	PUNCT
fcis-10369	158	1	techniques	technique	NOUN
fcis-10369	158	2	such	such	ADJ
fcis-10369	158	3	as	as	ADP
fcis-10369	158	4	max	max	PROPN
fcis-10369	158	5	pooling	pooling	NOUN
fcis-10369	158	6	,	,	PUNCT
fcis-10369	158	7	batch	batch	NOUN
fcis-10369	158	8	normalization	normalization	NOUN
fcis-10369	158	9	,	,	PUNCT
fcis-10369	158	10	and	and	CCONJ
fcis-10369	158	11	dropout	dropout	NOUN
fcis-10369	158	12	are	be	AUX
fcis-10369	158	13	utilized	utilize	VERB
fcis-10369	158	14	.	.	PUNCT
fcis-10369	159	1	the	the	DET
fcis-10369	159	2	specific	specific	ADJ
fcis-10369	159	3	parameter	parameter	NOUN
fcis-10369	159	4	settings	setting	NOUN
fcis-10369	159	5	are	be	AUX
fcis-10369	159	6	shown	show	VERB
fcis-10369	159	7	in	in	ADP
fcis-10369	159	8	table	table	NOUN
fcis-10369	159	9	3	3	NUM
fcis-10369	159	10	.	.	X
fcis-10369	159	11	4.4	4.4	NUM
fcis-10369	159	12	.	.	PUNCT
fcis-10369	160	1	evaluation	evaluation	NOUN
fcis-10369	160	2	metrics	metric	NOUN
fcis-10369	160	3	to	to	PART
fcis-10369	160	4	comprehensively	comprehensively	ADV
fcis-10369	160	5	evaluate	evaluate	VERB
fcis-10369	160	6	the	the	DET
fcis-10369	160	7	performance	performance	NOUN
fcis-10369	160	8	of	of	ADP
fcis-10369	160	9	the	the	DET
fcis-10369	160	10	msff	msff	NOUN
fcis-10369	160	11	model	model	NOUN
fcis-10369	160	12	,	,	PUNCT
fcis-10369	160	13	we	we	PRON
fcis-10369	160	14	conducted	conduct	VERB
fcis-10369	160	15	comparative	comparative	ADJ
fcis-10369	160	16	experiments	experiment	NOUN
fcis-10369	160	17	using	use	VERB
fcis-10369	160	18	multiple	multiple	ADJ
fcis-10369	160	19	metrics	metric	NOUN
fcis-10369	160	20	,	,	PUNCT
fcis-10369	160	21	including	include	VERB
fcis-10369	160	22	accuracy	accuracy	NOUN
fcis-10369	160	23	,	,	PUNCT
fcis-10369	160	24	precision	precision	NOUN
fcis-10369	160	25	,	,	PUNCT
fcis-10369	160	26	recall	recall	NOUN
fcis-10369	160	27	,	,	PUNCT
fcis-10369	160	28	false	false	ADJ
fcis-10369	160	29	positive	positive	ADJ
fcis-10369	160	30	rate	rate	NOUN
fcis-10369	160	31	(	(	PUNCT
fcis-10369	160	32	fpr	fpr	NOUN
fcis-10369	160	33	)	)	PUNCT
fcis-10369	160	34	,	,	PUNCT
fcis-10369	160	35	and	and	CCONJ
fcis-10369	160	36	f1	f1	PROPN
fcis-10369	160	37	score	score	NOUN
fcis-10369	160	38	.	.	PUNCT
fcis-10369	161	1	the	the	DET
fcis-10369	161	2	formulas	formula	NOUN
fcis-10369	161	3	for	for	ADP
fcis-10369	161	4	these	these	DET
fcis-10369	161	5	metrics	metric	NOUN
fcis-10369	161	6	are	be	AUX
fcis-10369	161	7	as	as	SCONJ
fcis-10369	161	8	follows	follow	VERB
fcis-10369	161	9	:	:	PUNCT
fcis-10369	161	10	tn	tn	PROPN
fcis-10369	161	11	tp	tp	PROPN
fcis-10369	161	12	accuracy	accuracy	PROPN
fcis-10369	161	13	tn	tn	PROPN
fcis-10369	161	14	tp	tp	PROPN
fcis-10369	161	15	fn	fn	PROPN
fcis-10369	161	16	fp	fp	PROPN
fcis-10369	161	17			PROPN
fcis-10369	161	18			PROPN
fcis-10369	161	19			PROPN
fcis-10369	161	20			PUNCT
fcis-10369	161	21			X
fcis-10369	161	22	(	(	PUNCT
fcis-10369	161	23	3	3	X
fcis-10369	161	24	)	)	PUNCT
fcis-10369	161	25	tp	tp	AUX
fcis-10369	161	26	precision	precision	NOUN
fcis-10369	161	27	tp	tp	ADP
fcis-10369	161	28	fp	fp	PROPN
fcis-10369	161	29			PROPN
fcis-10369	161	30			X
fcis-10369	161	31	(	(	PUNCT
fcis-10369	161	32	4	4	NUM
fcis-10369	161	33	)	)	PUNCT
fcis-10369	161	34	e	e	NOUN
fcis-10369	161	35	tp	tp	ADP
fcis-10369	161	36	r	r	NOUN
fcis-10369	161	37	call	call	NOUN
fcis-10369	161	38	tp	tp	ADP
fcis-10369	161	39	fn	fn	PROPN
fcis-10369	162	1			PROPN
fcis-10369	162	2			X
fcis-10369	162	3	(	(	PUNCT
fcis-10369	162	4	5	5	NUM
fcis-10369	162	5	)	)	PUNCT
fcis-10369	162	6	fp	fp	NOUN
fcis-10369	162	7	fpr	fpr	X
fcis-10369	162	8	fp	fp	PROPN
fcis-10369	162	9	tn	tn	PROPN
fcis-10369	162	10			PROPN
fcis-10369	162	11			PUNCT
fcis-10369	162	12	(	(	PUNCT
fcis-10369	162	13	6	6	NUM
fcis-10369	162	14	)	)	SYM
fcis-10369	162	15	2	2	NUM
fcis-10369	162	16	r	r	NOUN
fcis-10369	162	17	e	e	NOUN
fcis-10369	162	18	1	1	NUM
fcis-10369	162	19	r	r	NOUN
fcis-10369	162	20	e	e	NOUN
fcis-10369	162	21	p	p	NOUN
fcis-10369	162	22	ecision	ecision	NOUN
fcis-10369	162	23	r	r	NOUN
fcis-10369	162	24	call	call	NOUN
fcis-10369	162	25	f	f	PROPN
fcis-10369	162	26	score	score	NOUN
fcis-10369	162	27	p	p	PROPN
fcis-10369	162	28	ecision	ecision	NOUN
fcis-10369	162	29	r	r	NOUN
fcis-10369	162	30	call	call	NOUN
fcis-10369	162	31			PROPN
fcis-10369	163	1			PROPN
fcis-10369	163	2			VERB
fcis-10369	163	3			PRON
fcis-10369	163	4			PUNCT
fcis-10369	163	5	(	(	PUNCT
fcis-10369	163	6	7	7	NUM
fcis-10369	163	7	)	)	PUNCT
fcis-10369	163	8	where	where	SCONJ
fcis-10369	163	9	:	:	PUNCT
fcis-10369	163	10	tp	tp	X
fcis-10369	163	11	:	:	PUNCT
fcis-10369	163	12	true	true	ADJ
fcis-10369	163	13	positives	positive	NOUN
fcis-10369	163	14	(	(	PUNCT
fcis-10369	163	15	the	the	DET
fcis-10369	163	16	number	number	NOUN
fcis-10369	163	17	of	of	ADP
fcis-10369	163	18	correctly	correctly	ADV
fcis-10369	163	19	predicted	predict	VERB
fcis-10369	163	20	positive	positive	ADJ
fcis-10369	163	21	samples	sample	NOUN
fcis-10369	163	22	)	)	PUNCT
fcis-10369	163	23	;	;	PUNCT
fcis-10369	163	24	tn	tn	NOUN
fcis-10369	163	25	:	:	PUNCT
fcis-10369	164	1	true	true	ADJ
fcis-10369	164	2	negatives	negative	NOUN
fcis-10369	164	3	(	(	PUNCT
fcis-10369	164	4	the	the	DET
fcis-10369	164	5	number	number	NOUN
fcis-10369	164	6	of	of	ADP
fcis-10369	164	7	correctly	correctly	ADV
fcis-10369	164	8	predicted	predict	VERB
fcis-10369	164	9	negative	negative	ADJ
fcis-10369	164	10	samples	sample	NOUN
fcis-10369	164	11	)	)	PUNCT
fcis-10369	164	12	;	;	PUNCT
fcis-10369	164	13	fp	fp	X
fcis-10369	164	14	:	:	PUNCT
fcis-10369	164	15	false	false	ADJ
fcis-10369	164	16	positives	positive	NOUN
fcis-10369	164	17	(	(	PUNCT
fcis-10369	164	18	the	the	DET
fcis-10369	164	19	number	number	NOUN
fcis-10369	164	20	of	of	ADP
fcis-10369	164	21	incorrectly	incorrectly	ADV
fcis-10369	164	22	predicted	predict	VERB
fcis-10369	164	23	positive	positive	ADJ
fcis-10369	164	24	samples	sample	NOUN
fcis-10369	164	25	)	)	PUNCT
fcis-10369	164	26	;	;	PUNCT
fcis-10369	164	27	fn	fn	X
fcis-10369	164	28	:	:	PUNCT
fcis-10369	164	29	false	false	ADJ
fcis-10369	164	30	negatives	negative	NOUN
fcis-10369	164	31	(	(	PUNCT
fcis-10369	164	32	the	the	DET
fcis-10369	164	33	number	number	NOUN
fcis-10369	164	34	of	of	ADP
fcis-10369	164	35	incorrectly	incorrectly	ADV
fcis-10369	164	36	predicted	predict	VERB
fcis-10369	164	37	negative	negative	ADJ
fcis-10369	164	38	samples	sample	NOUN
fcis-10369	164	39	)	)	PUNCT
fcis-10369	164	40	4.5	4.5	NUM
fcis-10369	164	41	.	.	PUNCT
fcis-10369	165	1	binary	binary	ADJ
fcis-10369	165	2	classification	classification	NOUN
fcis-10369	165	3	results	result	VERB
fcis-10369	165	4	this	this	DET
fcis-10369	165	5	section	section	NOUN
fcis-10369	165	6	presents	present	VERB
fcis-10369	165	7	the	the	DET
fcis-10369	165	8	experiments	experiment	NOUN
fcis-10369	165	9	conducted	conduct	VERB
fcis-10369	165	10	on	on	ADP
fcis-10369	165	11	the	the	DET
fcis-10369	165	12	publicly	publicly	ADV
fcis-10369	165	13	available	available	ADJ
fcis-10369	165	14	network	network	NOUN
fcis-10369	165	15	intrusion	intrusion	NOUN
fcis-10369	165	16	detection	detection	NOUN
fcis-10369	165	17	datasets	dataset	NOUN
fcis-10369	165	18	,	,	PUNCT
fcis-10369	165	19	nslkdd	nslkdd	ADJ
fcis-10369	165	20	and	and	CCONJ
fcis-10369	165	21	unsw	unsw	NOUN
fcis-10369	165	22	-	-	PUNCT
fcis-10369	165	23	nb15	nb15	PROPN
fcis-10369	165	24	,	,	PUNCT
fcis-10369	165	25	for	for	ADP
fcis-10369	165	26	binary	binary	ADJ
fcis-10369	165	27	classification	classification	NOUN
fcis-10369	165	28	.	.	PUNCT
fcis-10369	166	1	to	to	PART
fcis-10369	166	2	validate	validate	VERB
fcis-10369	166	3	the	the	DET
fcis-10369	166	4	effectiveness	effectiveness	NOUN
fcis-10369	166	5	of	of	ADP
fcis-10369	166	6	the	the	DET
fcis-10369	166	7	proposed	propose	VERB
fcis-10369	166	8	model	model	NOUN
fcis-10369	166	9	,	,	PUNCT
fcis-10369	166	10	we	we	PRON
fcis-10369	166	11	compared	compare	VERB
fcis-10369	166	12	it	it	PRON
fcis-10369	166	13	with	with	ADP
fcis-10369	166	14	three	three	NUM
fcis-10369	166	15	classical	classical	ADJ
fcis-10369	166	16	machine	machine	NOUN
fcis-10369	166	17	learning	learning	NOUN
fcis-10369	166	18	algorithms	algorithm	NOUN
fcis-10369	166	19	,	,	PUNCT
fcis-10369	166	20	two	two	NUM
fcis-10369	166	21	individual	individual	ADJ
fcis-10369	166	22	deep	deep	ADJ
fcis-10369	166	23	learning	learning	NOUN
fcis-10369	166	24	models	model	NOUN
fcis-10369	166	25	,	,	PUNCT
fcis-10369	166	26	and	and	CCONJ
fcis-10369	166	27	two	two	NUM
fcis-10369	166	28	recently	recently	ADV
fcis-10369	166	29	proposed	propose	VERB
fcis-10369	166	30	deep	deep	ADJ
fcis-10369	166	31	learning	learning	NOUN
fcis-10369	166	32	models	model	NOUN
fcis-10369	166	33	.	.	PUNCT
fcis-10369	167	1	the	the	DET
fcis-10369	167	2	comparison	comparison	NOUN
fcis-10369	167	3	results	result	NOUN
fcis-10369	167	4	are	be	AUX
fcis-10369	167	5	shown	show	VERB
fcis-10369	167	6	in	in	ADP
fcis-10369	167	7	table	table	NOUN
fcis-10369	167	8	4	4	NUM
fcis-10369	167	9	.	.	PUNCT
fcis-10369	168	1	the	the	DET
fcis-10369	168	2	two	two	NUM
fcis-10369	168	3	deep	deep	ADJ
fcis-10369	168	4	learning	learning	NOUN
fcis-10369	168	5	models	model	NOUN
fcis-10369	168	6	are	be	AUX
fcis-10369	168	7	bcnndfs	bcnndfs	NOUN
fcis-10369	168	8	proposed	propose	VERB
fcis-10369	168	9	in	in	ADP
fcis-10369	168	10	reference	reference	NOUN
fcis-10369	168	11	[	[	X
fcis-10369	168	12	19	19	NUM
fcis-10369	168	13	]	]	PUNCT
fcis-10369	168	14	and	and	CCONJ
fcis-10369	168	15	semi	semi	ADJ
fcis-10369	168	16	-	-	ADJ
fcis-10369	168	17	gru	gru	NOUN
fcis-10369	168	18	proposed	propose	VERB
fcis-10369	168	19	in	in	ADP
fcis-10369	168	20	reference	reference	NOUN
fcis-10369	168	21	[	[	X
fcis-10369	168	22	9	9	NUM
fcis-10369	168	23	]	]	PUNCT
fcis-10369	168	24	.	.	PUNCT
fcis-10369	169	1	table	table	NOUN
fcis-10369	169	2	4	4	NUM
fcis-10369	169	3	.	.	PUNCT
fcis-10369	169	4	binary	binary	ADJ
fcis-10369	169	5	classification	classification	NOUN
fcis-10369	169	6	comparison	comparison	NOUN
fcis-10369	169	7	of	of	ADP
fcis-10369	169	8	models	model	NOUN
fcis-10369	169	9	model	model	VERB
fcis-10369	169	10	nsl	nsl	PROPN
fcis-10369	169	11	-	-	PUNCT
fcis-10369	169	12	kdd	kdd	PROPN
fcis-10369	169	13	unsw	unsw	PROPN
fcis-10369	169	14	-	-	PUNCT
fcis-10369	169	15	nb15	nb15	PROPN
fcis-10369	169	16	acc/%	acc/%	NOUN
fcis-10369	169	17	pre/%	pre/%	PROPN
fcis-10369	169	18	recall/%	recall/%	AUX
fcis-10369	169	19	f1/%	f1/%	PROPN
fcis-10369	169	20	acc/%	acc/%	NOUN
fcis-10369	169	21	pre/%	pre/%	PROPN
fcis-10369	169	22	recall/%	recall/%	AUX
fcis-10369	169	23	f1/%	f1/%	PROPN
fcis-10369	169	24	svm	svm	VERB
fcis-10369	169	25	87.33	87.33	NUM
fcis-10369	169	26	87.96	87.96	NUM
fcis-10369	169	27	87.33	87.33	NUM
fcis-10369	169	28	85.01	85.01	NUM
fcis-10369	169	29	62.00	62.00	NUM
fcis-10369	169	30	62.00	62.00	NUM
fcis-10369	169	31	60.00	60.00	NUM
fcis-10369	169	32	60.98	60.98	NUM
fcis-10369	169	33	rf	rf	NUM
fcis-10369	169	34	80.45	80.45	NUM
fcis-10369	169	35	97.05	97.05	NUM
fcis-10369	169	36	67.72	67.72	NUM
fcis-10369	169	37	70.77	70.77	NUM
fcis-10369	169	38	80.45	80.45	NUM
fcis-10369	169	39	97.05	97.05	NUM
fcis-10369	169	40	67.72	67.72	NUM
fcis-10369	169	41	79.77	79.77	NUM
fcis-10369	169	42	j48	j48	NOUN
fcis-10369	169	43	81.53	81.53	NUM
fcis-10369	169	44	97.14	97.14	NUM
fcis-10369	169	45	69.61	69.61	NUM
fcis-10369	169	46	81.10	81.10	NUM
fcis-10369	169	47	76.95	76.95	NUM
fcis-10369	169	48	70.05	70.05	NUM
fcis-10369	169	49	99.98	99.98	NUM
fcis-10369	169	50	82.69	82.69	NUM
fcis-10369	169	51	cnn	cnn	PROPN
fcis-10369	169	52	93.80	93.80	NUM
fcis-10369	169	53	98.80	98.80	NUM
fcis-10369	169	54	93.40	93.40	NUM
fcis-10369	169	55	96.02	96.02	NUM
fcis-10369	169	56	91.20	91.20	NUM
fcis-10369	169	57	87.53	87.53	NUM
fcis-10369	169	58	96.17	96.17	NUM
fcis-10369	169	59	91.59	91.59	NUM
fcis-10369	169	60	lstm	lstm	NOUN
fcis-10369	169	61	94.26	94.26	NUM
fcis-10369	169	62	99.05	99.05	NUM
fcis-10369	169	63	90.79	90.79	NUM
fcis-10369	169	64	94.74	94.74	NUM
fcis-10369	169	65	89.90	89.90	NUM
fcis-10369	169	66	88.90	88.90	NUM
fcis-10369	169	67	97.30	97.30	NUM
fcis-10369	169	68	92.90	92.90	NUM
fcis-10369	169	69	bcnn	bcnn	PROPN
fcis-10369	169	70	-	-	PUNCT
fcis-10369	169	71	dfs	dfs	PROPN
fcis-10369	169	72	90.14	90.14	NUM
fcis-10369	169	73	90.00	90.00	NUM
fcis-10369	169	74	90.00	90.00	NUM
fcis-10369	169	75	90.00	90.00	NUM
fcis-10369	169	76	89.26	89.26	NUM
fcis-10369	169	77	89.00	89.00	NUM
fcis-10369	169	78	89.00	89.00	NUM
fcis-10369	169	79	89.00	89.00	NUM
fcis-10369	169	80	semi	semi	NOUN
fcis-10369	169	81	-	-	NOUN
fcis-10369	169	82	gru	gru	NOUN
fcis-10369	169	83	92.18	92.18	NUM
fcis-10369	169	84	93.74	93.74	NUM
fcis-10369	169	85	92.43	92.43	NUM
fcis-10369	169	86	93.08	93.08	NUM
fcis-10369	169	87	88.11	88.11	NUM
fcis-10369	169	88	96.29	96.29	NUM
fcis-10369	169	89	81.56	81.56	NUM
fcis-10369	169	90	88.13	88.13	NUM
fcis-10369	169	91	msff	msff	NOUN
fcis-10369	169	92	99.50	99.50	NUM
fcis-10369	169	93	99.32	99.32	NUM
fcis-10369	169	94	99.59	99.59	NUM
fcis-10369	169	95	99.45	99.45	NUM
fcis-10369	169	96	94.73	94.73	NUM
fcis-10369	169	97	94.51	94.51	NUM
fcis-10369	169	98	91.20	91.20	NUM
fcis-10369	169	99	92.83	92.83	NUM
fcis-10369	169	100	from	from	ADP
fcis-10369	169	101	table	table	NOUN
fcis-10369	169	102	4	4	NUM
fcis-10369	169	103	,	,	PUNCT
fcis-10369	169	104	it	it	PRON
fcis-10369	169	105	can	can	AUX
fcis-10369	169	106	be	be	AUX
fcis-10369	169	107	observed	observe	VERB
fcis-10369	169	108	that	that	SCONJ
fcis-10369	169	109	the	the	DET
fcis-10369	169	110	proposed	propose	VERB
fcis-10369	169	111	model	model	NOUN
fcis-10369	169	112	in	in	ADP
fcis-10369	169	113	this	this	DET
fcis-10369	169	114	paper	paper	NOUN
fcis-10369	169	115	performs	perform	VERB
fcis-10369	169	116	well	well	ADV
fcis-10369	169	117	compared	compare	VERB
fcis-10369	169	118	to	to	ADP
fcis-10369	169	119	other	other	ADJ
fcis-10369	169	120	models	model	NOUN
fcis-10369	169	121	.	.	PUNCT
fcis-10369	170	1	it	it	PRON
fcis-10369	170	2	achieves	achieve	VERB
fcis-10369	170	3	an	an	DET
fcis-10369	170	4	accuracy	accuracy	NOUN
fcis-10369	170	5	of	of	ADP
fcis-10369	170	6	99.50	99.50	NUM
fcis-10369	170	7	%	%	NOUN
fcis-10369	170	8	,	,	PUNCT
fcis-10369	170	9	precision	precision	NOUN
fcis-10369	170	10	of	of	ADP
fcis-10369	170	11	99.32	99.32	NUM
fcis-10369	170	12	%	%	NOUN
fcis-10369	170	13	,	,	PUNCT
fcis-10369	170	14	recall	recall	NOUN
fcis-10369	170	15	of	of	ADP
fcis-10369	170	16	99.59	99.59	NUM
fcis-10369	170	17	%	%	NOUN
fcis-10369	170	18	,	,	PUNCT
fcis-10369	170	19	and	and	CCONJ
fcis-10369	170	20	f1	f1	ADJ
fcis-10369	170	21	score	score	NOUN
fcis-10369	170	22	of	of	ADP
fcis-10369	170	23	99.45	99.45	NUM
fcis-10369	170	24	%	%	NOUN
fcis-10369	170	25	.	.	PUNCT
fcis-10369	171	1	on	on	ADP
fcis-10369	171	2	the	the	DET
fcis-10369	171	3	other	other	ADJ
fcis-10369	171	4	hand	hand	NOUN
fcis-10369	171	5	,	,	PUNCT
fcis-10369	171	6	rf	rf	NOUN
fcis-10369	171	7	,	,	PUNCT
fcis-10369	171	8	svm	svm	ADJ
fcis-10369	171	9	,	,	PUNCT
fcis-10369	171	10	and	and	CCONJ
fcis-10369	171	11	j48	j48	PROPN
fcis-10369	171	12	exhibit	exhibit	VERB
fcis-10369	171	13	lower	low	ADJ
fcis-10369	171	14	accuracy	accuracy	NOUN
fcis-10369	171	15	with	with	ADP
fcis-10369	171	16	values	value	NOUN
fcis-10369	171	17	of	of	ADP
fcis-10369	171	18	80.45	80.45	NUM
fcis-10369	171	19	%	%	NOUN
fcis-10369	171	20	,	,	PUNCT
fcis-10369	171	21	87.33	87.33	NUM
fcis-10369	171	22	%	%	NOUN
fcis-10369	171	23	,	,	PUNCT
fcis-10369	171	24	and	and	CCONJ
fcis-10369	171	25	81.53	81.53	NUM
fcis-10369	171	26	%	%	NOUN
fcis-10369	171	27	,	,	PUNCT
fcis-10369	171	28	respectively	respectively	ADV
fcis-10369	171	29	.	.	PUNCT
fcis-10369	172	1	this	this	PRON
fcis-10369	172	2	is	be	AUX
fcis-10369	172	3	because	because	SCONJ
fcis-10369	172	4	they	they	PRON
fcis-10369	172	5	are	be	AUX
fcis-10369	172	6	unable	unable	ADJ
fcis-10369	172	7	to	to	PART
fcis-10369	172	8	fully	fully	ADV
fcis-10369	172	9	learn	learn	VERB
fcis-10369	172	10	the	the	DET
fcis-10369	172	11	data	data	NOUN
fcis-10369	172	12	features	feature	NOUN
fcis-10369	172	13	.	.	PUNCT
fcis-10369	173	1	when	when	SCONJ
fcis-10369	173	2	compared	compare	VERB
fcis-10369	173	3	to	to	ADP
fcis-10369	173	4	the	the	DET
fcis-10369	173	5	individual	individual	ADJ
fcis-10369	173	6	models	model	NOUN
fcis-10369	173	7	cnn	cnn	PROPN
fcis-10369	173	8	and	and	CCONJ
fcis-10369	173	9	lstm	lstm	PROPN
fcis-10369	173	10	,	,	PUNCT
fcis-10369	173	11	the	the	DET
fcis-10369	173	12	msff	msff	NOUN
fcis-10369	173	13	model	model	NOUN
fcis-10369	173	14	outperforms	outperform	VERB
fcis-10369	173	15	them	they	PRON
fcis-10369	173	16	in	in	ADP
fcis-10369	173	17	all	all	DET
fcis-10369	173	18	performance	performance	NOUN
fcis-10369	173	19	metrics	metric	NOUN
fcis-10369	173	20	except	except	SCONJ
fcis-10369	173	21	for	for	ADP
fcis-10369	173	22	a	a	DET
fcis-10369	173	23	slightly	slightly	ADV
fcis-10369	173	24	lower	low	ADJ
fcis-10369	173	25	recall	recall	NOUN
fcis-10369	173	26	in	in	ADP
fcis-10369	173	27	the	the	DET
fcis-10369	173	28	unsw	unsw	PROPN
fcis-10369	173	29	-	-	PUNCT
fcis-10369	173	30	nb15	nb15	PROPN
fcis-10369	173	31	dataset	dataset	NOUN
fcis-10369	173	32	,	,	PUNCT
fcis-10369	173	33	indicating	indicate	VERB
fcis-10369	173	34	the	the	DET
fcis-10369	173	35	msff	msff	NOUN
fcis-10369	173	36	model	model	NOUN
fcis-10369	173	37	's	's	PART
fcis-10369	173	38	ability	ability	NOUN
fcis-10369	173	39	to	to	PART
fcis-10369	173	40	effectively	effectively	ADV
fcis-10369	173	41	extract	extract	VERB
fcis-10369	173	42	network	network	NOUN
fcis-10369	173	43	data	data	NOUN
fcis-10369	173	44	features	feature	NOUN
fcis-10369	173	45	.	.	PUNCT
fcis-10369	174	1	compared	compare	VERB
fcis-10369	174	2	to	to	ADP
fcis-10369	174	3	the	the	DET
fcis-10369	174	4	semi	semi	ADJ
fcis-10369	174	5	-	-	ADJ
fcis-10369	174	6	gru	gru	NOUN
fcis-10369	174	7	model	model	NOUN
fcis-10369	174	8	,	,	PUNCT
fcis-10369	174	9	the	the	DET
fcis-10369	174	10	msff	msff	NOUN
fcis-10369	174	11	model	model	NOUN
fcis-10369	174	12	demonstrates	demonstrate	VERB
fcis-10369	174	13	an	an	DET
fcis-10369	174	14	improvement	improvement	NOUN
fcis-10369	174	15	of	of	ADP
fcis-10369	174	16	7.32	7.32	NUM
fcis-10369	174	17	%	%	NOUN
fcis-10369	174	18	in	in	ADP
fcis-10369	174	19	accuracy	accuracy	NOUN
fcis-10369	174	20	,	,	PUNCT
fcis-10369	174	21	5.58	5.58	NUM
fcis-10369	174	22	%	%	NOUN
fcis-10369	174	23	in	in	ADP
fcis-10369	174	24	precision	precision	NOUN
fcis-10369	174	25	,	,	PUNCT
fcis-10369	174	26	7.16	7.16	NUM
fcis-10369	174	27	%	%	NOUN
fcis-10369	174	28	in	in	ADP
fcis-10369	174	29	recall	recall	NOUN
fcis-10369	174	30	,	,	PUNCT
fcis-10369	174	31	and	and	CCONJ
fcis-10369	174	32	6.37	6.37	NUM
fcis-10369	174	33	%	%	NOUN
fcis-10369	174	34	in	in	ADP
fcis-10369	174	35	f1	f1	ADJ
fcis-10369	174	36	score	score	NOUN
fcis-10369	174	37	on	on	ADP
fcis-10369	174	38	the	the	DET
fcis-10369	174	39	nsl	nsl	NOUN
fcis-10369	174	40	-	-	PUNCT
fcis-10369	174	41	kdd	kdd	PROPN
fcis-10369	174	42	dataset	dataset	NOUN
fcis-10369	174	43	.	.	PUNCT
fcis-10369	175	1	similarly	similarly	ADV
fcis-10369	175	2	,	,	PUNCT
fcis-10369	175	3	on	on	ADP
fcis-10369	175	4	the	the	DET
fcis-10369	175	5	unsw	unsw	PROPN
fcis-10369	175	6	-	-	PUNCT
fcis-10369	175	7	nb15	nb15	PROPN
fcis-10369	175	8	dataset	dataset	NOUN
fcis-10369	175	9	,	,	PUNCT
fcis-10369	175	10	the	the	DET
fcis-10369	175	11	msff	msff	NOUN
fcis-10369	175	12	model	model	NOUN
fcis-10369	175	13	shows	show	VERB
fcis-10369	175	14	an	an	DET
fcis-10369	175	15	enhancement	enhancement	NOUN
fcis-10369	175	16	of	of	ADP
fcis-10369	175	17	6.62	6.62	NUM
fcis-10369	175	18	%	%	NOUN
fcis-10369	175	19	in	in	ADP
fcis-10369	175	20	accuracy	accuracy	NOUN
fcis-10369	175	21	,	,	PUNCT
fcis-10369	175	22	9.64	9.64	NUM
fcis-10369	175	23	%	%	NOUN
fcis-10369	175	24	in	in	ADP
fcis-10369	175	25	recall	recall	NOUN
fcis-10369	175	26	,	,	PUNCT
fcis-10369	175	27	and	and	CCONJ
fcis-10369	175	28	4.70	4.70	NUM
fcis-10369	175	29	%	%	NOUN
fcis-10369	175	30	in	in	ADP
fcis-10369	175	31	f1	f1	PROPN
fcis-10369	175	32	score	score	NOUN
fcis-10369	175	33	.	.	PUNCT
fcis-10369	176	1	overall	overall	ADV
fcis-10369	176	2	,	,	PUNCT
fcis-10369	176	3	the	the	DET
fcis-10369	176	4	msff	msff	NOUN
fcis-10369	176	5	model	model	NOUN
fcis-10369	176	6	exhibits	exhibit	VERB
fcis-10369	176	7	significant	significant	ADJ
fcis-10369	176	8	improvements	improvement	NOUN
fcis-10369	176	9	in	in	ADP
fcis-10369	176	10	all	all	DET
fcis-10369	176	11	the	the	DET
fcis-10369	176	12	evaluated	evaluated	ADJ
fcis-10369	176	13	metrics	metric	NOUN
fcis-10369	176	14	.	.	PUNCT
fcis-10369	177	1	4.6	4.6	NUM
fcis-10369	177	2	.	.	PUNCT
fcis-10369	178	1	multi	multi	ADJ
fcis-10369	178	2	-	-	ADJ
fcis-10369	178	3	classification	classification	ADJ
fcis-10369	178	4	results	result	NOUN
fcis-10369	178	5	this	this	DET
fcis-10369	178	6	section	section	NOUN
fcis-10369	178	7	presents	present	VERB
fcis-10369	178	8	the	the	DET
fcis-10369	178	9	experiments	experiment	NOUN
fcis-10369	178	10	conducted	conduct	VERB
fcis-10369	178	11	on	on	ADP
fcis-10369	178	12	the	the	DET
fcis-10369	178	13	publicly	publicly	ADV
fcis-10369	178	14	available	available	ADJ
fcis-10369	178	15	network	network	NOUN
fcis-10369	178	16	intrusion	intrusion	NOUN
fcis-10369	178	17	detection	detection	NOUN
fcis-10369	178	18	datasets	dataset	NOUN
fcis-10369	178	19	,	,	PUNCT
fcis-10369	178	20	nslkdd	nslkdd	ADJ
fcis-10369	178	21	and	and	CCONJ
fcis-10369	178	22	unsw	unsw	NOUN
fcis-10369	178	23	-	-	PUNCT
fcis-10369	178	24	nb15	nb15	PROPN
fcis-10369	178	25	,	,	PUNCT
fcis-10369	178	26	for	for	ADP
fcis-10369	178	27	multi	multi	ADJ
fcis-10369	178	28	-	-	ADJ
fcis-10369	178	29	class	class	ADJ
fcis-10369	178	30	classification	classification	NOUN
fcis-10369	178	31	.	.	PUNCT
fcis-10369	179	1	to	to	PART
fcis-10369	179	2	validate	validate	VERB
fcis-10369	179	3	the	the	DET
fcis-10369	179	4	superiority	superiority	NOUN
fcis-10369	179	5	of	of	ADP
fcis-10369	179	6	the	the	DET
fcis-10369	179	7	msff	msff	NOUN
fcis-10369	179	8	model	model	NOUN
fcis-10369	179	9	in	in	ADP
fcis-10369	179	10	improving	improve	VERB
fcis-10369	179	11	the	the	DET
fcis-10369	179	12	performance	performance	NOUN
fcis-10369	179	13	of	of	ADP
fcis-10369	179	14	minority	minority	NOUN
fcis-10369	179	15	classes	class	NOUN
fcis-10369	179	16	,	,	PUNCT
fcis-10369	179	17	we	we	PRON
fcis-10369	179	18	compared	compare	VERB
fcis-10369	179	19	129	129	NUM
fcis-10369	179	20	it	it	PRON
fcis-10369	179	21	with	with	ADP
fcis-10369	179	22	the	the	DET
fcis-10369	179	23	svm	svm	ADJ
fcis-10369	179	24	model	model	NOUN
fcis-10369	179	25	,	,	PUNCT
fcis-10369	179	26	the	the	DET
fcis-10369	179	27	id	id	PROPN
fcis-10369	179	28	-	-	PUNCT
fcis-10369	179	29	gan	gin	VERB
fcis-10369	179	30	model	model	NOUN
fcis-10369	179	31	proposed	propose	VERB
fcis-10369	179	32	in	in	ADP
fcis-10369	179	33	[	[	X
fcis-10369	179	34	19	19	NUM
fcis-10369	179	35	]	]	PUNCT
fcis-10369	179	36	,	,	PUNCT
fcis-10369	179	37	and	and	CCONJ
fcis-10369	179	38	the	the	DET
fcis-10369	179	39	stacon	stacon	PROPN
fcis-10369	179	40	-	-	PUNCT
fcis-10369	179	41	att	att	PROPN
fcis-10369	179	42	model	model	NOUN
fcis-10369	179	43	proposed	propose	VERB
fcis-10369	179	44	in	in	ADP
fcis-10369	179	45	[	[	X
fcis-10369	179	46	20	20	NUM
fcis-10369	179	47	]	]	PUNCT
fcis-10369	179	48	on	on	ADP
fcis-10369	179	49	the	the	DET
fcis-10369	179	50	nslkdd	nslkdd	ADJ
fcis-10369	179	51	dataset	dataset	NOUN
fcis-10369	179	52	.	.	PUNCT
fcis-10369	180	1	the	the	DET
fcis-10369	180	2	experimental	experimental	ADJ
fcis-10369	180	3	results	result	NOUN
fcis-10369	180	4	are	be	AUX
fcis-10369	180	5	shown	show	VERB
fcis-10369	180	6	in	in	ADP
fcis-10369	180	7	table	table	NOUN
fcis-10369	180	8	5	5	NUM
fcis-10369	180	9	.	.	PUNCT
fcis-10369	180	10	table	table	NOUN
fcis-10369	180	11	5	5	NUM
fcis-10369	180	12	.	.	PUNCT
fcis-10369	180	13	multi	multi	ADJ
fcis-10369	180	14	-	-	ADJ
fcis-10369	180	15	class	class	ADJ
fcis-10369	180	16	performance	performance	NOUN
fcis-10369	180	17	comparison	comparison	NOUN
fcis-10369	180	18	of	of	ADP
fcis-10369	180	19	different	different	ADJ
fcis-10369	180	20	models	model	NOUN
fcis-10369	180	21	on	on	ADP
fcis-10369	180	22	nsl	nsl	NOUN
fcis-10369	180	23	-	-	PUNCT
fcis-10369	180	24	kdd	kdd	PROPN
fcis-10369	180	25	category	category	NOUN
fcis-10369	180	26	evaluation	evaluation	NOUN
fcis-10369	180	27	metrics	metric	NOUN
fcis-10369	180	28	svm	svm	VERB
fcis-10369	180	29	id	id	ADJ
fcis-10369	180	30	-	-	PUNCT
fcis-10369	180	31	gan	gin	VERB
fcis-10369	180	32	stacon	stacon	NOUN
fcis-10369	180	33	-	-	PUNCT
fcis-10369	180	34	attn	attn	NOUN
fcis-10369	180	35	msff	msff	NOUN
fcis-10369	180	36	total	total	VERB
fcis-10369	180	37	accuracy/%	accuracy/%	ADP
fcis-10369	180	38	77.36	77.36	NUM
fcis-10369	180	39	-80.76	-80.76	PROPN
fcis-10369	180	40	99.50	99.50	NUM
fcis-10369	180	41	fpr/%	fpr/%	NOUN
fcis-10369	180	42	1.97	1.97	NUM
fcis-10369	180	43	-3.11	-3.11	NUM
fcis-10369	180	44	0.41	0.41	NUM
fcis-10369	180	45	normal	normal	ADJ
fcis-10369	180	46	precision/%	precision/%	ADP
fcis-10369	180	47	67.00	67.00	NUM
fcis-10369	180	48	75.10	75.10	NUM
fcis-10369	180	49	75.00	75.00	NUM
fcis-10369	180	50	99.32	99.32	NUM
fcis-10369	180	51	recall/%	recall/%	ADP
fcis-10369	180	52	98.00	98.00	NUM
fcis-10369	180	53	96.64	96.64	NUM
fcis-10369	180	54	97.00	97.00	NUM
fcis-10369	180	55	99.55	99.55	NUM
fcis-10369	180	56	f1/%	f1/%	NOUN
fcis-10369	180	57	80.00	80.00	NUM
fcis-10369	180	58	84.52	84.52	NUM
fcis-10369	180	59	85.00	85.00	NUM
fcis-10369	180	60	99.43	99.43	NUM
fcis-10369	180	61	dos	do	NOUN
fcis-10369	180	62	precision/%	precision/%	ADP
fcis-10369	180	63	97.00	97.00	NUM
fcis-10369	180	64	96.18	96.18	NUM
fcis-10369	180	65	97.00	97.00	NUM
fcis-10369	180	66	99.82	99.82	NUM
fcis-10369	180	67	recall/%	recall/%	NUM
fcis-10369	180	68	82.00	82.00	NUM
fcis-10369	180	69	83.70	83.70	NUM
fcis-10369	180	70	78.00	78.00	NUM
fcis-10369	180	71	99.88	99.88	NUM
fcis-10369	180	72	f1/%	f1/%	NOUN
fcis-10369	180	73	89.00	89.00	NUM
fcis-10369	180	74	89.50	89.50	NUM
fcis-10369	180	75	86.00	86.00	NUM
fcis-10369	180	76	99.85	99.85	NUM
fcis-10369	180	77	probe	probe	NOUN
fcis-10369	180	78	precision/%	precision/%	ADP
fcis-10369	180	79	85.00	85.00	NUM
fcis-10369	180	80	82.75	82.75	NUM
fcis-10369	180	81	70.00	70.00	NUM
fcis-10369	180	82	99.36	99.36	NUM
fcis-10369	180	83	recall/%	recall/%	ADP
fcis-10369	180	84	65.00	65.00	NUM
fcis-10369	180	85	84.01	84.01	NUM
fcis-10369	180	86	87.00	87.00	NUM
fcis-10369	180	87	99.19	99.19	NUM
fcis-10369	180	88	f1/%	f1/%	NOUN
fcis-10369	180	89	74.00	74.00	NUM
fcis-10369	180	90	83.38	83.38	NUM
fcis-10369	180	91	78.00	78.00	NUM
fcis-10369	180	92	99.27	99.27	NUM
fcis-10369	180	93	r2l	r2l	NOUN
fcis-10369	180	94	precision/%	precision/%	X
fcis-10369	180	95	95.00	95.00	NUM
fcis-10369	180	96	90.85	90.85	NUM
fcis-10369	180	97	93.00	93.00	NUM
fcis-10369	180	98	95.14	95.14	NUM
fcis-10369	180	99	recall/%	recall/%	NUM
fcis-10369	180	100	8.00	8.00	NUM
fcis-10369	180	101	34.60	34.60	NUM
fcis-10369	180	102	32.00	32.00	NUM
fcis-10369	180	103	89.62	89.62	NUM
fcis-10369	180	104	f1/%	f1/%	NOUN
fcis-10369	180	105	15.00	15.00	NUM
fcis-10369	180	106	50.12	50.12	NUM
fcis-10369	180	107	48.00	48.00	NUM
fcis-10369	180	108	92.30	92.30	NUM
fcis-10369	180	109	u2r	u2r	X
fcis-10369	180	110	precision/%	precision/%	X
fcis-10369	180	111	0.00	0.00	NUM
fcis-10369	180	112	62.00	62.00	NUM
fcis-10369	180	113	34.00	34.00	NUM
fcis-10369	180	114	52.63	52.63	NUM
fcis-10369	180	115	recall/%	recall/%	ADP
fcis-10369	180	116	0.00	0.00	NUM
fcis-10369	180	117	15.50	15.50	NUM
fcis-10369	180	118	7.00	7.00	NUM
fcis-10369	180	119	83.33	83.33	NUM
fcis-10369	180	120	f1/%	f1/%	NOUN
fcis-10369	180	121	0.00	0.00	NUM
fcis-10369	180	122	24.80	24.80	NUM
fcis-10369	180	123	12.00	12.00	NUM
fcis-10369	180	124	64.51	64.51	NUM
fcis-10369	180	125	table	table	NOUN
fcis-10369	180	126	5	5	NUM
fcis-10369	180	127	shows	show	VERB
fcis-10369	180	128	that	that	SCONJ
fcis-10369	180	129	the	the	DET
fcis-10369	180	130	overall	overall	ADJ
fcis-10369	180	131	accuracy	accuracy	NOUN
fcis-10369	180	132	of	of	ADP
fcis-10369	180	133	the	the	DET
fcis-10369	180	134	proposed	propose	VERB
fcis-10369	180	135	model	model	NOUN
fcis-10369	180	136	in	in	ADP
fcis-10369	180	137	this	this	DET
fcis-10369	180	138	paper	paper	NOUN
fcis-10369	180	139	reaches	reach	VERB
fcis-10369	180	140	99.50	99.50	NUM
fcis-10369	180	141	%	%	NOUN
fcis-10369	180	142	,	,	PUNCT
fcis-10369	180	143	with	with	ADP
fcis-10369	180	144	a	a	DET
fcis-10369	180	145	false	false	ADJ
fcis-10369	180	146	positive	positive	ADJ
fcis-10369	180	147	rate	rate	NOUN
fcis-10369	180	148	of	of	ADP
fcis-10369	180	149	0.41	0.41	NUM
fcis-10369	180	150	%	%	NOUN
fcis-10369	180	151	.	.	PUNCT
fcis-10369	181	1	compared	compare	VERB
fcis-10369	181	2	to	to	ADP
fcis-10369	181	3	other	other	ADJ
fcis-10369	181	4	models	model	NOUN
fcis-10369	181	5	,	,	PUNCT
fcis-10369	181	6	the	the	DET
fcis-10369	181	7	proposed	propose	VERB
fcis-10369	181	8	model	model	NOUN
fcis-10369	181	9	achieves	achieve	VERB
fcis-10369	181	10	the	the	DET
fcis-10369	181	11	highest	high	ADJ
fcis-10369	181	12	precision	precision	NOUN
fcis-10369	181	13	,	,	PUNCT
fcis-10369	181	14	recall	recall	NOUN
fcis-10369	181	15	,	,	PUNCT
fcis-10369	181	16	and	and	CCONJ
fcis-10369	181	17	f1	f1	NOUN
fcis-10369	181	18	score	score	NOUN
fcis-10369	181	19	in	in	ADP
fcis-10369	181	20	the	the	DET
fcis-10369	181	21	normal	normal	ADJ
fcis-10369	181	22	,	,	PUNCT
fcis-10369	181	23	dos	do	NOUN
fcis-10369	181	24	,	,	PUNCT
fcis-10369	181	25	probe	probe	NOUN
fcis-10369	181	26	,	,	PUNCT
fcis-10369	181	27	r2l	r2l	NOUN
fcis-10369	181	28	,	,	PUNCT
fcis-10369	181	29	and	and	CCONJ
fcis-10369	181	30	u2r	u2r	ADP
fcis-10369	181	31	categories	category	NOUN
fcis-10369	181	32	.	.	PUNCT
fcis-10369	182	1	the	the	DET
fcis-10369	182	2	precision	precision	NOUN
fcis-10369	182	3	rates	rate	NOUN
fcis-10369	182	4	are	be	AUX
fcis-10369	182	5	99.32	99.32	NUM
fcis-10369	182	6	%	%	NOUN
fcis-10369	182	7	,	,	PUNCT
fcis-10369	182	8	99.82	99.82	NUM
fcis-10369	182	9	%	%	NOUN
fcis-10369	182	10	,	,	PUNCT
fcis-10369	182	11	99.36	99.36	NUM
fcis-10369	182	12	%	%	NOUN
fcis-10369	182	13	,	,	PUNCT
fcis-10369	182	14	95.14	95.14	NUM
fcis-10369	182	15	%	%	NOUN
fcis-10369	182	16	,	,	PUNCT
fcis-10369	182	17	and	and	CCONJ
fcis-10369	182	18	52.63	52.63	NUM
fcis-10369	182	19	%	%	NOUN
fcis-10369	182	20	,	,	PUNCT
fcis-10369	182	21	respectively	respectively	ADV
fcis-10369	182	22	,	,	PUNCT
fcis-10369	182	23	while	while	SCONJ
fcis-10369	182	24	the	the	DET
fcis-10369	182	25	recall	recall	NOUN
fcis-10369	182	26	rates	rate	NOUN
fcis-10369	182	27	are	be	AUX
fcis-10369	182	28	99.55	99.55	NUM
fcis-10369	182	29	%	%	NOUN
fcis-10369	182	30	,	,	PUNCT
fcis-10369	182	31	99.88	99.88	NUM
fcis-10369	182	32	%	%	NOUN
fcis-10369	182	33	,	,	PUNCT
fcis-10369	182	34	99.19	99.19	NUM
fcis-10369	182	35	%	%	NOUN
fcis-10369	182	36	,	,	PUNCT
fcis-10369	182	37	89.62	89.62	NUM
fcis-10369	182	38	%	%	NOUN
fcis-10369	182	39	,	,	PUNCT
fcis-10369	182	40	and	and	CCONJ
fcis-10369	182	41	83.33	83.33	NUM
fcis-10369	182	42	%	%	NOUN
fcis-10369	182	43	,	,	PUNCT
fcis-10369	182	44	respectively	respectively	ADV
fcis-10369	182	45	.	.	PUNCT
fcis-10369	183	1	the	the	DET
fcis-10369	183	2	f1	f1	PROPN
fcis-10369	183	3	scores	score	NOUN
fcis-10369	183	4	are	be	AUX
fcis-10369	183	5	99.43	99.43	NUM
fcis-10369	183	6	%	%	NOUN
fcis-10369	183	7	,	,	PUNCT
fcis-10369	183	8	99.85	99.85	NUM
fcis-10369	183	9	%	%	NOUN
fcis-10369	183	10	,	,	PUNCT
fcis-10369	183	11	99.27	99.27	NUM
fcis-10369	183	12	%	%	NOUN
fcis-10369	183	13	,	,	PUNCT
fcis-10369	183	14	92.30	92.30	NUM
fcis-10369	183	15	%	%	NOUN
fcis-10369	183	16	,	,	PUNCT
fcis-10369	183	17	and	and	CCONJ
fcis-10369	183	18	64.51	64.51	NUM
fcis-10369	183	19	%	%	NOUN
fcis-10369	183	20	,	,	PUNCT
fcis-10369	183	21	respectively	respectively	ADV
fcis-10369	183	22	.	.	PUNCT
fcis-10369	184	1	based	base	VERB
fcis-10369	184	2	on	on	ADP
fcis-10369	184	3	the	the	DET
fcis-10369	184	4	comparison	comparison	NOUN
fcis-10369	184	5	results	result	VERB
fcis-10369	184	6	,	,	PUNCT
fcis-10369	184	7	the	the	DET
fcis-10369	184	8	msff	msff	NOUN
fcis-10369	184	9	model	model	NOUN
fcis-10369	184	10	demonstrates	demonstrate	VERB
fcis-10369	184	11	excellent	excellent	ADJ
fcis-10369	184	12	performance	performance	NOUN
fcis-10369	184	13	in	in	ADP
fcis-10369	184	14	recognizing	recognize	VERB
fcis-10369	184	15	minority	minority	NOUN
fcis-10369	184	16	classes	class	NOUN
fcis-10369	184	17	.	.	PUNCT
fcis-10369	185	1	table	table	NOUN
fcis-10369	185	2	6	6	NUM
fcis-10369	185	3	.	.	PUNCT
fcis-10369	186	1	comparison	comparison	NOUN
fcis-10369	186	2	of	of	ADP
fcis-10369	186	3	f1	f1	NOUN
fcis-10369	186	4	scores	score	NOUN
fcis-10369	186	5	(	(	PUNCT
fcis-10369	186	6	%	%	INTJ
fcis-10369	186	7	)	)	PUNCT
fcis-10369	186	8	of	of	ADP
fcis-10369	186	9	different	different	ADJ
fcis-10369	186	10	models	model	NOUN
fcis-10369	186	11	on	on	ADP
fcis-10369	186	12	unsw	unsw	PROPN
fcis-10369	186	13	-	-	PUNCT
fcis-10369	186	14	nb15	nb15	PROPN
fcis-10369	186	15	category	category	NOUN
fcis-10369	186	16	rf	rf	NOUN
fcis-10369	186	17	alexnet	alexnet	ADJ
fcis-10369	186	18	cnn	cnn	PROPN
fcis-10369	186	19	lstm	lstm	PROPN
fcis-10369	186	20	msff	msff	NOUN
fcis-10369	186	21	normal	normal	ADJ
fcis-10369	186	22	85.86	85.86	NUM
fcis-10369	186	23	86.85	86.85	NUM
fcis-10369	186	24	85.20	85.20	NUM
fcis-10369	186	25	88.41	88.41	NUM
fcis-10369	186	26	91.85	91.85	NUM
fcis-10369	186	27	generic	generic	ADJ
fcis-10369	186	28	98.98	98.98	NUM
fcis-10369	186	29	95.26	95.26	NUM
fcis-10369	186	30	97.31	97.31	NUM
fcis-10369	186	31	98.48	98.48	NUM
fcis-10369	186	32	98.89	98.89	NUM
fcis-10369	186	33	exploits	exploit	VERB
fcis-10369	186	34	63.37	63.37	NUM
fcis-10369	186	35	66.95	66.95	NUM
fcis-10369	186	36	67.73	67.73	NUM
fcis-10369	186	37	58.19	58.19	NUM
fcis-10369	186	38	74.23	74.23	NUM
fcis-10369	186	39	fuzzers	fuzzer	NOUN
fcis-10369	186	40	28.13	28.13	NUM
fcis-10369	186	41	26.59	26.59	NUM
fcis-10369	186	42	34.68	34.68	NUM
fcis-10369	186	43	28.31	28.31	NUM
fcis-10369	186	44	65.02	65.02	NUM
fcis-10369	187	1	dos	do	VERB
fcis-10369	187	2	19.04	19.04	NUM
fcis-10369	187	3	24.16	24.16	NUM
fcis-10369	187	4	5.33	5.33	NUM
fcis-10369	187	5	6.10	6.10	NUM
fcis-10369	187	6	31.76	31.76	NUM
fcis-10369	187	7	reconnaissance	reconnaissance	NOUN
fcis-10369	187	8	58.11	58.11	NUM
fcis-10369	187	9	53.37	53.37	NUM
fcis-10369	187	10	75.21	75.21	NUM
fcis-10369	187	11	56.65	56.65	NUM
fcis-10369	187	12	83.91	83.91	NUM
fcis-10369	187	13	analysis	analysis	NOUN
fcis-10369	187	14	6.93	6.93	NUM
fcis-10369	187	15	4.88	4.88	NUM
fcis-10369	187	16	2.09	2.09	NUM
fcis-10369	187	17	2.91	2.91	NUM
fcis-10369	187	18	15.29	15.29	NUM
fcis-10369	187	19	backdoor	backdoor	NOUN
fcis-10369	187	20	4.33	4.33	NUM
fcis-10369	187	21	3.62	3.62	NUM
fcis-10369	187	22	2.53	2.53	NUM
fcis-10369	187	23	2.90	2.90	NUM
fcis-10369	187	24	15.81	15.81	NUM
fcis-10369	187	25	shellcode	shellcode	NOUN
fcis-10369	187	26	32.59	32.59	NUM
fcis-10369	187	27	23.84	23.84	NUM
fcis-10369	187	28	24.64	24.64	NUM
fcis-10369	187	29	18.41	18.41	NUM
fcis-10369	187	30	61.26	61.26	NUM
fcis-10369	187	31	worms	worm	NOUN
fcis-10369	187	32	6.05	6.05	NUM
fcis-10369	187	33	3.17	3.17	NUM
fcis-10369	187	34	1.58	1.58	NUM
fcis-10369	187	35	1.95	1.95	NUM
fcis-10369	187	36	60.21	60.21	NUM
fcis-10369	187	37	to	to	PART
fcis-10369	187	38	verify	verify	VERB
fcis-10369	187	39	the	the	DET
fcis-10369	187	40	applicability	applicability	NOUN
fcis-10369	187	41	of	of	ADP
fcis-10369	187	42	the	the	DET
fcis-10369	187	43	msff	msff	NOUN
fcis-10369	187	44	model	model	NOUN
fcis-10369	187	45	,	,	PUNCT
fcis-10369	187	46	more	more	ADV
fcis-10369	187	47	indepth	indepth	ADJ
fcis-10369	187	48	comparative	comparative	ADJ
fcis-10369	187	49	experiments	experiment	NOUN
fcis-10369	187	50	were	be	AUX
fcis-10369	187	51	conducted	conduct	VERB
fcis-10369	187	52	on	on	ADP
fcis-10369	187	53	the	the	DET
fcis-10369	187	54	unsw	unsw	PROPN
fcis-10369	187	55	-	-	PUNCT
fcis-10369	187	56	nb15	nb15	PROPN
fcis-10369	187	57	dataset	dataset	NOUN
fcis-10369	187	58	.	.	PUNCT
fcis-10369	188	1	the	the	DET
fcis-10369	188	2	f1	f1	PROPN
fcis-10369	188	3	score	score	NOUN
fcis-10369	188	4	,	,	PUNCT
fcis-10369	188	5	which	which	PRON
fcis-10369	188	6	combines	combine	VERB
fcis-10369	188	7	precision	precision	NOUN
fcis-10369	188	8	and	and	CCONJ
fcis-10369	188	9	recall	recall	NOUN
fcis-10369	188	10	,	,	PUNCT
fcis-10369	188	11	effectively	effectively	ADV
fcis-10369	188	12	evaluates	evaluate	VERB
fcis-10369	188	13	the	the	DET
fcis-10369	188	14	performance	performance	NOUN
fcis-10369	188	15	of	of	ADP
fcis-10369	188	16	the	the	DET
fcis-10369	188	17	model	model	NOUN
fcis-10369	188	18	.	.	PUNCT
fcis-10369	189	1	table	table	NOUN
fcis-10369	189	2	6	6	NUM
fcis-10369	189	3	presents	present	VERB
fcis-10369	189	4	the	the	DET
fcis-10369	189	5	comparison	comparison	NOUN
fcis-10369	189	6	of	of	ADP
fcis-10369	189	7	f1	f1	NOUN
fcis-10369	189	8	scores	score	NOUN
fcis-10369	189	9	for	for	ADP
fcis-10369	189	10	different	different	ADJ
fcis-10369	189	11	categories	category	NOUN
fcis-10369	189	12	in	in	ADP
fcis-10369	189	13	the	the	DET
fcis-10369	189	14	dataset	dataset	NOUN
fcis-10369	189	15	under	under	ADP
fcis-10369	189	16	different	different	ADJ
fcis-10369	189	17	models	model	NOUN
fcis-10369	189	18	.	.	PUNCT
fcis-10369	190	1	from	from	ADP
fcis-10369	190	2	table	table	NOUN
fcis-10369	190	3	6	6	NUM
fcis-10369	190	4	,	,	PUNCT
fcis-10369	190	5	it	it	PRON
fcis-10369	190	6	can	can	AUX
fcis-10369	190	7	be	be	AUX
fcis-10369	190	8	observed	observe	VERB
fcis-10369	190	9	that	that	SCONJ
fcis-10369	190	10	the	the	DET
fcis-10369	190	11	categories	category	NOUN
fcis-10369	190	12	analysis	analysis	NOUN
fcis-10369	190	13	,	,	PUNCT
fcis-10369	190	14	backdoor	backdoor	NOUN
fcis-10369	190	15	,	,	PUNCT
fcis-10369	190	16	and	and	CCONJ
fcis-10369	190	17	worms	worm	NOUN
fcis-10369	190	18	have	have	VERB
fcis-10369	190	19	relatively	relatively	ADV
fcis-10369	190	20	smaller	small	ADJ
fcis-10369	190	21	f1	f1	NOUN
fcis-10369	190	22	scores	score	NOUN
fcis-10369	190	23	compared	compare	VERB
fcis-10369	190	24	to	to	ADP
fcis-10369	190	25	other	other	ADJ
fcis-10369	190	26	categories	category	NOUN
fcis-10369	190	27	like	like	ADP
fcis-10369	190	28	normal	normal	ADJ
fcis-10369	190	29	and	and	CCONJ
fcis-10369	190	30	generic	generic	ADJ
fcis-10369	190	31	.	.	PUNCT
fcis-10369	191	1	compared	compare	VERB
fcis-10369	191	2	to	to	ADP
fcis-10369	191	3	other	other	ADJ
fcis-10369	191	4	models	model	NOUN
fcis-10369	191	5	,	,	PUNCT
fcis-10369	191	6	the	the	DET
fcis-10369	191	7	msff	msff	NOUN
fcis-10369	191	8	model	model	NOUN
fcis-10369	191	9	shows	show	VERB
fcis-10369	191	10	significant	significant	ADJ
fcis-10369	191	11	improvement	improvement	NOUN
fcis-10369	191	12	in	in	ADP
fcis-10369	191	13	f1	f1	NOUN
fcis-10369	191	14	scores	score	NOUN
fcis-10369	191	15	for	for	ADP
fcis-10369	191	16	the	the	DET
fcis-10369	191	17	analysis	analysis	NOUN
fcis-10369	191	18	,	,	PUNCT
fcis-10369	191	19	backdoor	backdoor	NOUN
fcis-10369	191	20	,	,	PUNCT
fcis-10369	191	21	worms	worm	NOUN
fcis-10369	191	22	,	,	PUNCT
fcis-10369	191	23	and	and	CCONJ
fcis-10369	191	24	shellcode	shellcode	PROPN
fcis-10369	191	25	categories	category	NOUN
fcis-10369	191	26	,	,	PUNCT
fcis-10369	191	27	ranging	range	VERB
fcis-10369	191	28	from	from	ADP
fcis-10369	191	29	8.36	8.36	NUM
fcis-10369	191	30	%	%	NOUN
fcis-10369	191	31	to	to	ADP
fcis-10369	191	32	13.20	13.20	NUM
fcis-10369	191	33	%	%	NOUN
fcis-10369	191	34	,	,	PUNCT
fcis-10369	191	35	11.48	11.48	NUM
fcis-10369	191	36	%	%	NOUN
fcis-10369	191	37	to	to	ADP
fcis-10369	191	38	13.28	13.28	NUM
fcis-10369	191	39	%	%	NOUN
fcis-10369	191	40	,	,	PUNCT
fcis-10369	191	41	54.16	54.16	NUM
fcis-10369	191	42	%	%	NOUN
fcis-10369	191	43	to	to	ADP
fcis-10369	191	44	58.63	58.63	NUM
fcis-10369	191	45	%	%	NOUN
fcis-10369	191	46	,	,	PUNCT
fcis-10369	191	47	and	and	CCONJ
fcis-10369	191	48	28.67	28.67	NUM
fcis-10369	191	49	%	%	NOUN
fcis-10369	191	50	to	to	ADP
fcis-10369	191	51	42.85	42.85	NUM
fcis-10369	191	52	%	%	NOUN
fcis-10369	191	53	,	,	PUNCT
fcis-10369	191	54	respectively	respectively	ADV
fcis-10369	191	55	.	.	PUNCT
fcis-10369	192	1	this	this	PRON
fcis-10369	192	2	indicates	indicate	VERB
fcis-10369	192	3	that	that	SCONJ
fcis-10369	192	4	the	the	DET
fcis-10369	192	5	msff	msff	NOUN
fcis-10369	192	6	model	model	NOUN
fcis-10369	192	7	effectively	effectively	ADV
fcis-10369	192	8	enhances	enhance	VERB
fcis-10369	192	9	the	the	DET
fcis-10369	192	10	detection	detection	NOUN
fcis-10369	192	11	performance	performance	NOUN
fcis-10369	192	12	for	for	ADP
fcis-10369	192	13	minority	minority	NOUN
fcis-10369	192	14	class	class	NOUN
fcis-10369	192	15	samples	sample	NOUN
fcis-10369	192	16	,	,	PUNCT
fcis-10369	192	17	demonstrating	demonstrate	VERB
fcis-10369	192	18	the	the	DET
fcis-10369	192	19	applicability	applicability	NOUN
fcis-10369	192	20	of	of	ADP
fcis-10369	192	21	this	this	DET
fcis-10369	192	22	approach	approach	NOUN
fcis-10369	192	23	.	.	PUNCT
fcis-10369	193	1	5	5	X
fcis-10369	193	2	.	.	X
fcis-10369	193	3	summary	summary	NOUN
fcis-10369	193	4	in	in	ADP
fcis-10369	193	5	response	response	NOUN
fcis-10369	193	6	to	to	ADP
fcis-10369	193	7	the	the	DET
fcis-10369	193	8	insufficient	insufficient	ADJ
fcis-10369	193	9	feature	feature	NOUN
fcis-10369	193	10	learning	learning	NOUN
fcis-10369	193	11	,	,	PUNCT
fcis-10369	193	12	representation	representation	NOUN
fcis-10369	193	13	capability	capability	NOUN
fcis-10369	193	14	,	,	PUNCT
fcis-10369	193	15	and	and	CCONJ
fcis-10369	193	16	generalization	generalization	NOUN
fcis-10369	193	17	ability	ability	NOUN
fcis-10369	193	18	of	of	ADP
fcis-10369	193	19	existing	exist	VERB
fcis-10369	193	20	intrusion	intrusion	NOUN
fcis-10369	193	21	detection	detection	NOUN
fcis-10369	193	22	models	model	NOUN
fcis-10369	193	23	,	,	PUNCT
fcis-10369	193	24	a	a	DET
fcis-10369	193	25	novel	novel	ADJ
fcis-10369	193	26	approach	approach	NOUN
fcis-10369	193	27	is	be	AUX
fcis-10369	193	28	proposed	propose	VERB
fcis-10369	193	29	in	in	ADP
fcis-10369	193	30	this	this	DET
fcis-10369	193	31	work	work	NOUN
fcis-10369	193	32	.	.	PUNCT
fcis-10369	194	1	it	it	PRON
fcis-10369	194	2	combines	combine	VERB
fcis-10369	194	3	multi	multi	ADJ
fcis-10369	194	4	-	-	ADJ
fcis-10369	194	5	scale	scale	ADJ
fcis-10369	194	6	1d	1d	NUM
fcis-10369	194	7	-	-	PUNCT
fcis-10369	194	8	cnn	cnn	NOUN
fcis-10369	194	9	and	and	CCONJ
fcis-10369	194	10	bilstm	bilstm	NOUN
fcis-10369	194	11	to	to	PART
fcis-10369	194	12	extract	extract	VERB
fcis-10369	194	13	fused	fuse	VERB
fcis-10369	194	14	features	feature	NOUN
fcis-10369	194	15	from	from	ADP
fcis-10369	194	16	the	the	DET
fcis-10369	194	17	data	datum	NOUN
fcis-10369	194	18	,	,	PUNCT
fcis-10369	194	19	while	while	SCONJ
fcis-10369	194	20	incorporating	incorporate	VERB
fcis-10369	194	21	residual	residual	ADJ
fcis-10369	194	22	connections	connection	NOUN
fcis-10369	194	23	with	with	ADP
fcis-10369	194	24	identity	identity	NOUN
fcis-10369	194	25	mapping	mapping	NOUN
fcis-10369	194	26	to	to	PART
fcis-10369	194	27	enhance	enhance	VERB
fcis-10369	194	28	model	model	NOUN
fcis-10369	194	29	performance	performance	NOUN
fcis-10369	194	30	.	.	PUNCT
fcis-10369	195	1	moreover	moreover	ADV
fcis-10369	195	2	,	,	PUNCT
fcis-10369	195	3	considering	consider	VERB
fcis-10369	195	4	the	the	DET
fcis-10369	195	5	problem	problem	NOUN
fcis-10369	195	6	of	of	ADP
fcis-10369	195	7	data	datum	NOUN
fcis-10369	195	8	imbalance	imbalance	NOUN
fcis-10369	195	9	in	in	ADP
fcis-10369	195	10	existing	exist	VERB
fcis-10369	195	11	public	public	ADJ
fcis-10369	195	12	network	network	NOUN
fcis-10369	195	13	intrusion	intrusion	NOUN
fcis-10369	195	14	datasets	dataset	NOUN
fcis-10369	195	15	,	,	PUNCT
fcis-10369	195	16	the	the	DET
fcis-10369	195	17	wgan	wgan	VERB
fcis-10369	195	18	-	-	PUNCT
fcis-10369	195	19	gp	gp	NOUN
fcis-10369	195	20	algorithm	algorithm	NOUN
fcis-10369	195	21	is	be	AUX
fcis-10369	195	22	employed	employ	VERB
fcis-10369	195	23	to	to	PART
fcis-10369	195	24	improve	improve	VERB
fcis-10369	195	25	the	the	DET
fcis-10369	195	26	detection	detection	NOUN
fcis-10369	195	27	performance	performance	NOUN
fcis-10369	195	28	for	for	ADP
fcis-10369	195	29	minority	minority	NOUN
fcis-10369	195	30	class	class	NOUN
fcis-10369	195	31	samples	sample	NOUN
fcis-10369	195	32	.	.	PUNCT
fcis-10369	196	1	the	the	DET
fcis-10369	196	2	proposed	propose	VERB
fcis-10369	196	3	msff	msff	NOUN
fcis-10369	196	4	model	model	NOUN
fcis-10369	196	5	is	be	AUX
fcis-10369	196	6	validated	validate	VERB
fcis-10369	196	7	on	on	ADP
fcis-10369	196	8	the	the	DET
fcis-10369	196	9	nsl	nsl	NOUN
fcis-10369	196	10	-	-	PUNCT
fcis-10369	196	11	kdd	kdd	PROPN
fcis-10369	196	12	and	and	CCONJ
fcis-10369	196	13	unsw	unsw	PROPN
fcis-10369	196	14	-	-	PUNCT
fcis-10369	196	15	nb15	nb15	PROPN
fcis-10369	196	16	datasets	dataset	NOUN
fcis-10369	196	17	.	.	PUNCT
fcis-10369	197	1	the	the	DET
fcis-10369	197	2	results	result	NOUN
fcis-10369	197	3	demonstrate	demonstrate	VERB
fcis-10369	197	4	that	that	SCONJ
fcis-10369	197	5	the	the	DET
fcis-10369	197	6	msff	msff	NOUN
fcis-10369	197	7	model	model	NOUN
fcis-10369	197	8	outperforms	outperform	VERB
fcis-10369	197	9	several	several	ADJ
fcis-10369	197	10	classical	classical	ADJ
fcis-10369	197	11	machine	machine	NOUN
fcis-10369	197	12	learning	learning	NOUN
fcis-10369	197	13	methods	method	NOUN
fcis-10369	197	14	and	and	CCONJ
fcis-10369	197	15	popular	popular	ADJ
fcis-10369	197	16	deep	deep	ADJ
fcis-10369	197	17	learning	learning	NOUN
fcis-10369	197	18	approaches	approach	NOUN
fcis-10369	197	19	in	in	ADP
fcis-10369	197	20	both	both	CCONJ
fcis-10369	197	21	binary	binary	ADJ
fcis-10369	197	22	130	130	NUM
fcis-10369	197	23	and	and	CCONJ
fcis-10369	197	24	multi	multi	ADJ
fcis-10369	197	25	-	-	ADJ
fcis-10369	197	26	class	class	ADJ
fcis-10369	197	27	experiments	experiment	NOUN
fcis-10369	197	28	,	,	PUNCT
fcis-10369	197	29	achieving	achieve	VERB
fcis-10369	197	30	high	high	ADJ
fcis-10369	197	31	accuracy	accuracy	NOUN
fcis-10369	197	32	,	,	PUNCT
fcis-10369	197	33	precision	precision	NOUN
fcis-10369	197	34	,	,	PUNCT
fcis-10369	197	35	recall	recall	NOUN
fcis-10369	197	36	,	,	PUNCT
fcis-10369	197	37	and	and	CCONJ
fcis-10369	197	38	f1	f1	PROPN
fcis-10369	197	39	scores	score	NOUN
fcis-10369	197	40	.	.	PUNCT
fcis-10369	198	1	in	in	ADP
fcis-10369	198	2	future	future	ADJ
fcis-10369	198	3	work	work	NOUN
fcis-10369	198	4	,	,	PUNCT
fcis-10369	198	5	further	further	ADJ
fcis-10369	198	6	exploration	exploration	NOUN
fcis-10369	198	7	will	will	AUX
fcis-10369	198	8	be	be	AUX
fcis-10369	198	9	conducted	conduct	VERB
fcis-10369	198	10	to	to	PART
fcis-10369	198	11	investigate	investigate	VERB
fcis-10369	198	12	algorithms	algorithm	NOUN
fcis-10369	198	13	that	that	PRON
fcis-10369	198	14	are	be	AUX
fcis-10369	198	15	better	well	ADV
fcis-10369	198	16	suited	suited	ADJ
fcis-10369	198	17	to	to	PART
fcis-10369	198	18	address	address	VERB
fcis-10369	198	19	data	datum	NOUN
fcis-10369	198	20	imbalance	imbalance	NOUN
fcis-10369	198	21	and	and	CCONJ
fcis-10369	198	22	enhance	enhance	VERB
fcis-10369	198	23	the	the	DET
fcis-10369	198	24	detection	detection	NOUN
fcis-10369	198	25	performance	performance	NOUN
fcis-10369	198	26	for	for	ADP
fcis-10369	198	27	anomalous	anomalous	ADJ
fcis-10369	198	28	traffic	traffic	NOUN
fcis-10369	198	29	.	.	PUNCT
fcis-10369	199	1	additionally	additionally	ADV
fcis-10369	199	2	,	,	PUNCT
fcis-10369	199	3	the	the	DET
fcis-10369	199	4	use	use	NOUN
fcis-10369	199	5	of	of	ADP
fcis-10369	199	6	updated	update	VERB
fcis-10369	199	7	and	and	CCONJ
fcis-10369	199	8	more	more	ADV
fcis-10369	199	9	realistic	realistic	ADJ
fcis-10369	199	10	datasets	dataset	NOUN
fcis-10369	199	11	will	will	AUX
fcis-10369	199	12	be	be	AUX
fcis-10369	199	13	explored	explore	VERB
fcis-10369	199	14	for	for	ADP
fcis-10369	199	15	validation	validation	NOUN
fcis-10369	199	16	purposes	purpose	NOUN
fcis-10369	199	17	.	.	PUNCT
fcis-10369	200	1	acknowledgments	acknowledgment	NOUN
fcis-10369	200	2	sichuan	sichuan	PROPN
fcis-10369	200	3	science	science	PROPN
fcis-10369	200	4	and	and	CCONJ
fcis-10369	200	5	technology	technology	NOUN
fcis-10369	200	6	program	program	NOUN
fcis-10369	200	7	project	project	NOUN
fcis-10369	200	8	(	(	PUNCT
fcis-10369	200	9	2020yfg0151	2020yfg0151	NUM
fcis-10369	200	10	)	)	PUNCT
fcis-10369	200	11	;	;	PUNCT
fcis-10369	200	12	key	key	ADJ
fcis-10369	200	13	laboratory	laboratory	NOUN
fcis-10369	200	14	project	project	NOUN
fcis-10369	200	15	of	of	ADP
fcis-10369	200	16	enterprise	enterprise	NOUN
fcis-10369	200	17	informatization	informatization	NOUN
fcis-10369	200	18	and	and	CCONJ
fcis-10369	200	19	iot	iot	PROPN
fcis-10369	200	20	measurement	measurement	NOUN
fcis-10369	200	21	and	and	CCONJ
fcis-10369	200	22	control	control	NOUN
fcis-10369	200	23	technology	technology	PROPN
fcis-10369	200	24	in	in	ADP
fcis-10369	200	25	sichuan	sichuan	PROPN
fcis-10369	200	26	province	province	PROPN
fcis-10369	200	27	(	(	PUNCT
fcis-10369	200	28	2021wzy01	2021wzy01	NUM
fcis-10369	200	29	)	)	PUNCT
fcis-10369	200	30	.	.	PUNCT
fcis-10369	201	1	references	reference	NOUN
fcis-10369	201	2	[	[	X
fcis-10369	201	3	1	1	NUM
fcis-10369	201	4	]	]	X
fcis-10369	201	5	wazirali	wazirali	PROPN
fcis-10369	201	6	r.	r.	PROPN
fcis-10369	201	7	an	an	DET
fcis-10369	201	8	improved	improved	ADJ
fcis-10369	201	9	intrusion	intrusion	NOUN
fcis-10369	201	10	detection	detection	NOUN
fcis-10369	201	11	system	system	NOUN
fcis-10369	201	12	based	base	VERB
fcis-10369	201	13	on	on	ADP
fcis-10369	201	14	knn	knn	PROPN
fcis-10369	201	15	hyperparameter	hyperparameter	PROPN
fcis-10369	201	16	tuning	tuning	NOUN
fcis-10369	201	17	and	and	CCONJ
fcis-10369	201	18	cross	cross	NOUN
fcis-10369	201	19	-	-	PROPN
fcis-10369	201	20	validation[j	validation[j	ADJ
fcis-10369	201	21	]	]	PUNCT
fcis-10369	201	22	.	.	PUNCT
fcis-10369	202	1	arabian	arabian	ADJ
fcis-10369	202	2	journal	journal	PROPN
fcis-10369	202	3	for	for	ADP
fcis-10369	202	4	science	science	NOUN
fcis-10369	202	5	and	and	CCONJ
fcis-10369	202	6	engineering	engineering	NOUN
fcis-10369	202	7	,	,	PUNCT
fcis-10369	202	8	2020	2020	NUM
fcis-10369	202	9	,	,	PUNCT
fcis-10369	202	10	45(12	45(12	NUM
fcis-10369	202	11	):	):	PUNCT
fcis-10369	202	12	1085910873	1085910873	NUM
fcis-10369	202	13	.	.	PUNCT
fcis-10369	203	1	[	[	X
fcis-10369	203	2	2	2	X
fcis-10369	203	3	]	]	X
fcis-10369	203	4	liu	liu	PROPN
fcis-10369	203	5	g	g	PROPN
fcis-10369	203	6	,	,	PUNCT
fcis-10369	203	7	zhao	zhao	PROPN
fcis-10369	203	8	h	h	PROPN
fcis-10369	203	9	,	,	PUNCT
fcis-10369	203	10	fan	fan	PROPN
fcis-10369	203	11	f	f	PROPN
fcis-10369	203	12	,	,	PUNCT
fcis-10369	203	13	et	et	PROPN
fcis-10369	203	14	al	al	PROPN
fcis-10369	203	15	.	.	PUNCT
fcis-10369	204	1	an	an	DET
fcis-10369	204	2	enhanced	enhance	VERB
fcis-10369	204	3	intrusion	intrusion	NOUN
fcis-10369	204	4	detection	detection	NOUN
fcis-10369	204	5	model	model	NOUN
fcis-10369	204	6	based	base	VERB
fcis-10369	204	7	on	on	ADP
fcis-10369	204	8	improved	improved	ADJ
fcis-10369	204	9	knn	knn	PROPN
fcis-10369	204	10	in	in	ADP
fcis-10369	204	11	wsns[j	wsns[j	PROPN
fcis-10369	204	12	]	]	PUNCT
fcis-10369	204	13	.	.	PUNCT
fcis-10369	205	1	sensors	sensor	NOUN
fcis-10369	205	2	,	,	PUNCT
fcis-10369	205	3	2022	2022	NUM
fcis-10369	205	4	,	,	PUNCT
fcis-10369	205	5	22	22	NUM
fcis-10369	205	6	(	(	PUNCT
fcis-10369	205	7	4	4	NUM
fcis-10369	205	8	):	):	PUNCT
fcis-10369	205	9	1407	1407	NUM
fcis-10369	205	10	.	.	PUNCT
fcis-10369	206	1	[	[	X
fcis-10369	206	2	3	3	NUM
fcis-10369	206	3	]	]	X
fcis-10369	206	4	mohammadi	mohammadi	PROPN
fcis-10369	206	5	m	m	PROPN
fcis-10369	206	6	,	,	PUNCT
fcis-10369	206	7	rashid	rashid	PROPN
fcis-10369	206	8	t	t	PROPN
fcis-10369	206	9	a	a	PROPN
fcis-10369	206	10	,	,	PUNCT
fcis-10369	206	11	karim	karim	PROPN
fcis-10369	206	12	s	s	PROPN
fcis-10369	206	13	h	h	PROPN
fcis-10369	206	14	t	t	PROPN
fcis-10369	206	15	,	,	PUNCT
fcis-10369	206	16	et	et	PROPN
fcis-10369	206	17	al	al	PROPN
fcis-10369	206	18	.	.	PUNCT
fcis-10369	207	1	a	a	DET
fcis-10369	207	2	comprehensive	comprehensive	ADJ
fcis-10369	207	3	survey	survey	NOUN
fcis-10369	207	4	and	and	CCONJ
fcis-10369	207	5	taxonomy	taxonomy	NOUN
fcis-10369	207	6	of	of	ADP
fcis-10369	207	7	the	the	DET
fcis-10369	207	8	svm	svm	ADJ
fcis-10369	207	9	-	-	PUNCT
fcis-10369	207	10	based	base	VERB
fcis-10369	207	11	intrusion	intrusion	NOUN
fcis-10369	207	12	detection	detection	NOUN
fcis-10369	207	13	systems[j	systems[j	PROPN
fcis-10369	207	14	]	]	PUNCT
fcis-10369	207	15	.	.	PUNCT
fcis-10369	208	1	journal	journal	PROPN
fcis-10369	208	2	of	of	ADP
fcis-10369	208	3	network	network	NOUN
fcis-10369	208	4	and	and	CCONJ
fcis-10369	208	5	computer	computer	NOUN
fcis-10369	208	6	applications	application	NOUN
fcis-10369	208	7	,	,	PUNCT
fcis-10369	208	8	2021	2021	NUM
fcis-10369	208	9	,	,	PUNCT
fcis-10369	208	10	178	178	NUM
fcis-10369	208	11	:	:	SYM
fcis-10369	208	12	102983	102983	NUM
fcis-10369	208	13	.	.	PUNCT
fcis-10369	209	1	[	[	X
fcis-10369	209	2	4	4	X
fcis-10369	209	3	]	]	PUNCT
fcis-10369	209	4	wang	wang	PROPN
fcis-10369	209	5	h	h	PROPN
fcis-10369	209	6	,	,	PUNCT
fcis-10369	209	7	gu	gu	PROPN
fcis-10369	209	8	j	j	PROPN
fcis-10369	209	9	,	,	PUNCT
fcis-10369	209	10	wang	wang	PROPN
fcis-10369	209	11	s.	s.	PROPN
fcis-10369	209	12	an	an	DET
fcis-10369	209	13	effective	effective	ADJ
fcis-10369	209	14	intrusion	intrusion	NOUN
fcis-10369	209	15	detection	detection	NOUN
fcis-10369	209	16	framework	framework	NOUN
fcis-10369	209	17	based	base	VERB
fcis-10369	209	18	on	on	ADP
fcis-10369	209	19	svm	svm	PROPN
fcis-10369	209	20	with	with	ADP
fcis-10369	209	21	feature	feature	NOUN
fcis-10369	209	22	augmentation[j	augmentation[j	PROPN
fcis-10369	209	23	]	]	PUNCT
fcis-10369	209	24	.	.	PUNCT
fcis-10369	210	1	knowledge	knowledge	NOUN
fcis-10369	210	2	-	-	PUNCT
fcis-10369	210	3	based	base	VERB
fcis-10369	210	4	systems	system	NOUN
fcis-10369	210	5	,	,	PUNCT
fcis-10369	210	6	2017	2017	NUM
fcis-10369	210	7	,	,	PUNCT
fcis-10369	210	8	136	136	NUM
fcis-10369	210	9	:	:	PUNCT
fcis-10369	210	10	130	130	NUM
fcis-10369	210	11	-	-	SYM
fcis-10369	210	12	139	139	NUM
fcis-10369	210	13	.	.	PUNCT
fcis-10369	211	1	[	[	X
fcis-10369	211	2	5	5	NUM
fcis-10369	211	3	]	]	X
fcis-10369	211	4	tabash	tabash	NOUN
fcis-10369	211	5	m	m	PROPN
fcis-10369	211	6	,	,	PUNCT
fcis-10369	211	7	abd	abd	PROPN
fcis-10369	211	8	allah	allah	PROPN
fcis-10369	211	9	m	m	PROPN
fcis-10369	211	10	,	,	PUNCT
fcis-10369	211	11	tawfik	tawfik	PROPN
fcis-10369	211	12	b.	b.	PROPN
fcis-10369	211	13	intrusion	intrusion	PROPN
fcis-10369	211	14	detection	detection	NOUN
fcis-10369	211	15	model	model	NOUN
fcis-10369	211	16	using	use	VERB
fcis-10369	211	17	naive	naive	ADJ
fcis-10369	211	18	bayes	baye	NOUN
fcis-10369	211	19	and	and	CCONJ
fcis-10369	211	20	deep	deep	ADJ
fcis-10369	211	21	learning	learn	VERB
fcis-10369	211	22	technique[j	technique[j	PROPN
fcis-10369	211	23	]	]	PUNCT
fcis-10369	211	24	.	.	PUNCT
fcis-10369	212	1	int	int	NOUN
fcis-10369	212	2	.	.	PUNCT
fcis-10369	213	1	arab	arab	PROPN
fcis-10369	213	2	j.	j.	PROPN
fcis-10369	213	3	inf	inf	PROPN
fcis-10369	213	4	.	.	PUNCT
fcis-10369	213	5	technol	technol	PROPN
fcis-10369	213	6	.	.	PROPN
fcis-10369	213	7	,	,	PUNCT
fcis-10369	213	8	2020	2020	NUM
fcis-10369	213	9	,	,	PUNCT
fcis-10369	213	10	17(2	17(2	NUM
fcis-10369	213	11	):	):	PUNCT
fcis-10369	213	12	215	215	NUM
fcis-10369	213	13	-	-	SYM
fcis-10369	213	14	224	224	NUM
fcis-10369	213	15	.	.	PUNCT
fcis-10369	214	1	[	[	X
fcis-10369	214	2	6	6	NUM
fcis-10369	214	3	]	]	PUNCT
fcis-10369	214	4	balyan	balyan	NOUN
fcis-10369	214	5	a	a	DET
fcis-10369	214	6	k	k	NOUN
fcis-10369	214	7	,	,	PUNCT
fcis-10369	214	8	ahuja	ahuja	PROPN
fcis-10369	214	9	s	s	PROPN
fcis-10369	214	10	,	,	PUNCT
fcis-10369	214	11	lilhore	lilhore	PROPN
fcis-10369	214	12	u	u	PROPN
fcis-10369	214	13	k	k	PROPN
fcis-10369	214	14	,	,	PUNCT
fcis-10369	214	15	et	et	PROPN
fcis-10369	214	16	al	al	PROPN
fcis-10369	214	17	.	.	PUNCT
fcis-10369	215	1	a	a	DET
fcis-10369	215	2	hybrid	hybrid	ADJ
fcis-10369	215	3	intrusion	intrusion	NOUN
fcis-10369	215	4	detection	detection	NOUN
fcis-10369	215	5	model	model	NOUN
fcis-10369	215	6	using	use	VERB
fcis-10369	215	7	ega	ega	NOUN
fcis-10369	215	8	-	-	NOUN
fcis-10369	215	9	pso	pso	NOUN
fcis-10369	215	10	and	and	CCONJ
fcis-10369	215	11	improved	improve	VERB
fcis-10369	215	12	random	random	ADJ
fcis-10369	215	13	forest	forest	NOUN
fcis-10369	215	14	method[j	method[j	NOUN
fcis-10369	215	15	]	]	PUNCT
fcis-10369	215	16	.	.	PUNCT
fcis-10369	216	1	sensors	sensor	NOUN
fcis-10369	216	2	,	,	PUNCT
fcis-10369	216	3	2022	2022	NUM
fcis-10369	216	4	,	,	PUNCT
fcis-10369	216	5	22(16	22(16	NUM
fcis-10369	216	6	):	):	PUNCT
fcis-10369	216	7	5986	5986	NUM
fcis-10369	216	8	.	.	PUNCT
fcis-10369	217	1	[	[	X
fcis-10369	217	2	7	7	X
fcis-10369	217	3	]	]	X
fcis-10369	217	4	anton	anton	NOUN
fcis-10369	217	5	s	s	PROPN
fcis-10369	217	6	d	d	X
fcis-10369	217	7	d	d	PROPN
fcis-10369	217	8	,	,	PUNCT
fcis-10369	217	9	sinha	sinha	PROPN
fcis-10369	217	10	s	s	PROPN
fcis-10369	217	11	,	,	PUNCT
fcis-10369	217	12	schotten	schotten	ADJ
fcis-10369	217	13	h	h	PROPN
fcis-10369	217	14	d.	d.	PROPN
fcis-10369	217	15	anomaly	anomaly	PROPN
fcis-10369	217	16	-	-	PUNCT
fcis-10369	217	17	based	base	VERB
fcis-10369	217	18	intrusion	intrusion	NOUN
fcis-10369	217	19	detection	detection	NOUN
fcis-10369	217	20	in	in	ADP
fcis-10369	217	21	industrial	industrial	ADJ
fcis-10369	217	22	data	datum	NOUN
fcis-10369	217	23	with	with	ADP
fcis-10369	217	24	svm	svm	ADJ
fcis-10369	217	25	and	and	CCONJ
fcis-10369	217	26	random	random	ADJ
fcis-10369	217	27	forests[c]//2019	forests[c]//2019	PUNCT
fcis-10369	217	28	international	international	ADJ
fcis-10369	217	29	conference	conference	NOUN
fcis-10369	217	30	on	on	ADP
fcis-10369	217	31	software	software	NOUN
fcis-10369	217	32	,	,	PUNCT
fcis-10369	217	33	telecommunications	telecommunications	NOUN
fcis-10369	217	34	and	and	CCONJ
fcis-10369	217	35	computer	computer	NOUN
fcis-10369	217	36	networks	network	NOUN
fcis-10369	217	37	(	(	PUNCT
fcis-10369	217	38	softcom	softcom	PROPN
fcis-10369	217	39	)	)	PUNCT
fcis-10369	217	40	.	.	PUNCT
fcis-10369	218	1	ieee	ieee	PROPN
fcis-10369	218	2	,	,	PUNCT
fcis-10369	218	3	2019	2019	NUM
fcis-10369	218	4	:	:	PUNCT
fcis-10369	218	5	1	1	NUM
fcis-10369	218	6	-	-	SYM
fcis-10369	218	7	6	6	NUM
fcis-10369	218	8	.	.	PUNCT
fcis-10369	219	1	[	[	X
fcis-10369	219	2	8	8	NUM
fcis-10369	219	3	]	]	X
fcis-10369	219	4	shu	shu	PROPN
fcis-10369	219	5	hao	hao	PROPN
fcis-10369	219	6	,	,	PUNCT
fcis-10369	219	7	wang	wang	PROPN
fcis-10369	219	8	chen	chen	PROPN
fcis-10369	219	9	,	,	PUNCT
fcis-10369	219	10	shi	shi	PROPN
fcis-10369	219	11	yan	yan	PROPN
fcis-10369	219	12	.	.	PROPN
fcis-10369	219	13	intrusion	intrusion	NOUN
fcis-10369	219	14	detection	detection	NOUN
fcis-10369	219	15	based	base	VERB
fcis-10369	219	16	on	on	ADP
fcis-10369	219	17	bilstm	bilstm	NOUN
fcis-10369	219	18	and	and	CCONJ
fcis-10369	219	19	attention	attention	NOUN
fcis-10369	219	20	mechanism[j	mechanism[j	PROPN
fcis-10369	219	21	]	]	PUNCT
fcis-10369	219	22	.	.	PUNCT
fcis-10369	220	1	computer	computer	NOUN
fcis-10369	220	2	engineering	engineering	NOUN
fcis-10369	220	3	and	and	CCONJ
fcis-10369	220	4	design	design	NOUN
fcis-10369	220	5	,	,	PUNCT
fcis-10369	220	6	2020	2020	NUM
fcis-10369	220	7	,	,	PUNCT
fcis-10369	220	8	41(11	41(11	NUM
fcis-10369	220	9	):	):	PUNCT
fcis-10369	220	10	3042	3042	NUM
fcis-10369	220	11	-	-	SYM
fcis-10369	220	12	3046	3046	NUM
fcis-10369	220	13	.	.	PUNCT
fcis-10369	221	1	[	[	X
fcis-10369	221	2	9	9	NUM
fcis-10369	221	3	]	]	SYM
fcis-10369	221	4	li	li	PROPN
fcis-10369	221	5	haitao	haitao	PROPN
fcis-10369	221	6	,	,	PUNCT
fcis-10369	221	7	wang	wang	PROPN
fcis-10369	221	8	ruimin	ruimin	PROPN
fcis-10369	221	9	,	,	PUNCT
fcis-10369	221	10	dong	dong	PROPN
fcis-10369	221	11	weiyu	weiyu	PROPN
fcis-10369	221	12	,	,	PUNCT
fcis-10369	221	13	jiang	jiang	PROPN
fcis-10369	221	14	liehui	liehui	PROPN
fcis-10369	221	15	.	.	PUNCT
fcis-10369	222	1	a	a	DET
fcis-10369	222	2	semisupervised	semisupervise	VERB
fcis-10369	222	3	network	network	NOUN
fcis-10369	222	4	traffic	traffic	NOUN
fcis-10369	222	5	anomaly	anomaly	NOUN
fcis-10369	222	6	detection	detection	NOUN
fcis-10369	222	7	method	method	NOUN
fcis-10369	222	8	based	base	VERB
fcis-10369	222	9	on	on	ADP
fcis-10369	222	10	gru[j	gru[j	PROPN
fcis-10369	222	11	]	]	PUNCT
fcis-10369	222	12	.	.	PUNCT
fcis-10369	223	1	computer	computer	NOUN
fcis-10369	223	2	science	science	NOUN
fcis-10369	223	3	,	,	PUNCT
fcis-10369	223	4	2023	2023	NUM
fcis-10369	223	5	,	,	PUNCT
fcis-10369	223	6	50(03	50(03	NUM
fcis-10369	223	7	):	):	PUNCT
fcis-10369	223	8	380-390.kumar	380-390.kumar	PROPN
fcis-10369	223	9	j	j	PROPN
fcis-10369	223	10	,	,	PUNCT
fcis-10369	223	11	goomer	goomer	PROPN
fcis-10369	223	12	r	r	PROPN
fcis-10369	223	13	,	,	PUNCT
fcis-10369	223	14	singh	singh	PROPN
fcis-10369	223	15	a	a	DET
fcis-10369	223	16	k.	k.	PROPN
fcis-10369	223	17	long	long	PROPN
fcis-10369	223	18	short	short	ADJ
fcis-10369	223	19	term	term	NOUN
fcis-10369	223	20	memory	memory	NOUN
fcis-10369	223	21	recurrent	recurrent	NOUN
fcis-10369	223	22	neural	neural	ADJ
fcis-10369	223	23	network	network	NOUN
fcis-10369	223	24	(	(	PUNCT
fcis-10369	223	25	lstm	lstm	PROPN
fcis-10369	223	26	-	-	PUNCT
fcis-10369	223	27	rnn	rnn	PROPN
fcis-10369	223	28	)	)	PUNCT
fcis-10369	223	29	based	base	VERB
fcis-10369	223	30	workload	workload	NOUN
fcis-10369	223	31	forecasting	forecasting	NOUN
fcis-10369	223	32	model	model	NOUN
fcis-10369	223	33	for	for	ADP
fcis-10369	223	34	cloud	cloud	NOUN
fcis-10369	223	35	datacenters[j	datacenters[j	PROPN
fcis-10369	223	36	]	]	PUNCT
fcis-10369	223	37	.	.	PUNCT
fcis-10369	224	1	procedia	procedia	PROPN
fcis-10369	224	2	computer	computer	NOUN
fcis-10369	224	3	science	science	NOUN
fcis-10369	224	4	,	,	PUNCT
fcis-10369	224	5	2018	2018	NUM
fcis-10369	224	6	,	,	PUNCT
fcis-10369	224	7	125	125	NUM
fcis-10369	224	8	:	:	SYM
fcis-10369	224	9	676	676	NUM
fcis-10369	224	10	-	-	SYM
fcis-10369	224	11	682	682	NUM
fcis-10369	224	12	.	.	PUNCT
fcis-10369	225	1	[	[	X
fcis-10369	225	2	10	10	NUM
fcis-10369	225	3	]	]	X
fcis-10369	225	4	yu	yu	PROPN
fcis-10369	225	5	x	x	PROPN
fcis-10369	225	6	,	,	PUNCT
fcis-10369	225	7	li	li	PROPN
fcis-10369	225	8	t	t	PROPN
fcis-10369	225	9	,	,	PUNCT
fcis-10369	225	10	hu	hu	PROPN
fcis-10369	225	11	a.	a.	PROPN
fcis-10369	225	12	time	time	PROPN
fcis-10369	225	13	-	-	PUNCT
fcis-10369	225	14	series	series	NOUN
fcis-10369	225	15	network	network	NOUN
fcis-10369	225	16	anomaly	anomaly	NOUN
fcis-10369	225	17	detection	detection	NOUN
fcis-10369	225	18	based	base	VERB
fcis-10369	225	19	on	on	ADP
fcis-10369	225	20	behaviour	behaviour	NOUN
fcis-10369	225	21	characteristics[c	characteristics[c	PROPN
fcis-10369	225	22	]	]	PUNCT
fcis-10369	225	23	.	.	PUNCT
fcis-10369	226	1	in	in	ADP
fcis-10369	226	2	2020	2020	NUM
fcis-10369	226	3	ieee	ieee	NOUN
fcis-10369	226	4	6th	6th	ADJ
fcis-10369	226	5	international	international	ADJ
fcis-10369	226	6	conference	conference	NOUN
fcis-10369	226	7	on	on	ADP
fcis-10369	226	8	computer	computer	NOUN
fcis-10369	226	9	and	and	CCONJ
fcis-10369	226	10	communications	communication	NOUN
fcis-10369	226	11	(	(	PUNCT
fcis-10369	226	12	iccc	iccc	PROPN
fcis-10369	226	13	)	)	PUNCT
fcis-10369	226	14	.	.	PUNCT
fcis-10369	227	1	chengdu	chengdu	PROPN
fcis-10369	227	2	,	,	PUNCT
fcis-10369	227	3	china	china	PROPN
fcis-10369	227	4	:	:	PUNCT
fcis-10369	227	5	ieee	ieee	NOUN
fcis-10369	227	6	,	,	PUNCT
fcis-10369	227	7	2020	2020	NUM
fcis-10369	227	8	:	:	PUNCT
fcis-10369	227	9	568	568	NUM
fcis-10369	227	10	-	-	SYM
fcis-10369	227	11	572	572	NUM
fcis-10369	227	12	.	.	PUNCT
fcis-10369	228	1	[	[	X
fcis-10369	228	2	11	11	NUM
fcis-10369	228	3	]	]	X
fcis-10369	228	4	khan	khan	PROPN
fcis-10369	228	5	m	m	PROPN
fcis-10369	228	6	a.	a.	NOUN
fcis-10369	228	7	hcrnnids	hcrnnid	NOUN
fcis-10369	228	8	:	:	PUNCT
fcis-10369	228	9	hybrid	hybrid	ADJ
fcis-10369	228	10	convolutional	convolutional	ADJ
fcis-10369	228	11	recurrent	recurrent	ADJ
fcis-10369	228	12	neural	neural	ADJ
fcis-10369	228	13	network	network	NOUN
fcis-10369	228	14	-	-	PUNCT
fcis-10369	228	15	based	base	VERB
fcis-10369	228	16	network	network	NOUN
fcis-10369	228	17	intrusion	intrusion	NOUN
fcis-10369	228	18	detection	detection	NOUN
fcis-10369	228	19	system[j	system[j	NOUN
fcis-10369	228	20	]	]	PUNCT
fcis-10369	228	21	.	.	PUNCT
fcis-10369	229	1	processes	process	NOUN
fcis-10369	229	2	,	,	PUNCT
fcis-10369	229	3	2021	2021	NUM
fcis-10369	229	4	,	,	PUNCT
fcis-10369	229	5	9(5	9(5	NUM
fcis-10369	229	6	):	):	PUNCT
fcis-10369	229	7	834	834	NUM
fcis-10369	229	8	.	.	PUNCT
fcis-10369	230	1	[	[	X
fcis-10369	230	2	12	12	NUM
fcis-10369	230	3	]	]	PUNCT
fcis-10369	230	4	rodda	rodda	PROPN
fcis-10369	230	5	s	s	PROPN
fcis-10369	230	6	,	,	PUNCT
fcis-10369	230	7	erothi	erothi	ADJ
fcis-10369	230	8	u	u	PROPN
fcis-10369	230	9	s	s	PROPN
fcis-10369	230	10	r.	r.	PROPN
fcis-10369	230	11	class	class	PROPN
fcis-10369	230	12	imbalance	imbalance	NOUN
fcis-10369	230	13	problem	problem	NOUN
fcis-10369	230	14	in	in	ADP
fcis-10369	230	15	the	the	DET
fcis-10369	230	16	network	network	NOUN
fcis-10369	230	17	intrusion	intrusion	PROPN
fcis-10369	230	18	detection	detection	NOUN
fcis-10369	230	19	systems[c]//2016	systems[c]//2016	PROPN
fcis-10369	230	20	international	international	ADJ
fcis-10369	230	21	conference	conference	NOUN
fcis-10369	230	22	on	on	ADP
fcis-10369	230	23	electrical	electrical	ADJ
fcis-10369	230	24	,	,	PUNCT
fcis-10369	230	25	electronics	electronic	NOUN
fcis-10369	230	26	,	,	PUNCT
fcis-10369	230	27	and	and	CCONJ
fcis-10369	230	28	optimization	optimization	NOUN
fcis-10369	230	29	techniques	technique	NOUN
fcis-10369	230	30	(	(	PUNCT
fcis-10369	230	31	iceeot	iceeot	PROPN
fcis-10369	230	32	)	)	PUNCT
fcis-10369	230	33	.	.	PUNCT
fcis-10369	231	1	ieee	ieee	PROPN
fcis-10369	231	2	,	,	PUNCT
fcis-10369	231	3	2016	2016	NUM
fcis-10369	231	4	:	:	PUNCT
fcis-10369	231	5	2685	2685	NUM
fcis-10369	231	6	-	-	SYM
fcis-10369	231	7	2688	2688	NUM
fcis-10369	231	8	.	.	PUNCT
fcis-10369	232	1	[	[	X
fcis-10369	232	2	13	13	NUM
fcis-10369	232	3	]	]	X
fcis-10369	232	4	zhang	zhang	PROPN
fcis-10369	232	5	h	h	PROPN
fcis-10369	232	6	,	,	PUNCT
fcis-10369	232	7	huang	huang	PROPN
fcis-10369	232	8	l	l	PROPN
fcis-10369	232	9	,	,	PUNCT
fcis-10369	232	10	wu	wu	PROPN
fcis-10369	232	11	c	c	PROPN
fcis-10369	232	12	q	q	PROPN
fcis-10369	232	13	,	,	PUNCT
fcis-10369	232	14	et	et	PROPN
fcis-10369	232	15	al	al	PROPN
fcis-10369	232	16	.	.	PUNCT
fcis-10369	233	1	an	an	DET
fcis-10369	233	2	effective	effective	ADJ
fcis-10369	233	3	convolutional	convolutional	ADJ
fcis-10369	233	4	neural	neural	ADJ
fcis-10369	233	5	network	network	NOUN
fcis-10369	233	6	based	base	VERB
fcis-10369	233	7	on	on	ADP
fcis-10369	233	8	smote	smote	ADJ
fcis-10369	233	9	and	and	CCONJ
fcis-10369	233	10	gaussian	gaussian	ADJ
fcis-10369	233	11	mixture	mixture	NOUN
fcis-10369	233	12	model	model	NOUN
fcis-10369	233	13	for	for	ADP
fcis-10369	233	14	intrusion	intrusion	NOUN
fcis-10369	233	15	detection	detection	NOUN
fcis-10369	233	16	in	in	ADP
fcis-10369	233	17	imbalanced	imbalanced	ADJ
fcis-10369	233	18	dataset[j	dataset[j	NOUN
fcis-10369	233	19	]	]	PUNCT
fcis-10369	233	20	.	.	PUNCT
fcis-10369	234	1	computer	computer	NOUN
fcis-10369	234	2	networks	network	NOUN
fcis-10369	234	3	,	,	PUNCT
fcis-10369	234	4	2020	2020	NUM
fcis-10369	234	5	,	,	PUNCT
fcis-10369	234	6	177	177	NUM
fcis-10369	234	7	:	:	SYM
fcis-10369	234	8	107315	107315	NUM
fcis-10369	234	9	.	.	PUNCT
fcis-10369	235	1	[	[	X
fcis-10369	235	2	14	14	NUM
fcis-10369	235	3	]	]	X
fcis-10369	235	4	jiang	jiang	PROPN
fcis-10369	235	5	k	k	PROPN
fcis-10369	235	6	,	,	PUNCT
fcis-10369	235	7	wang	wang	PROPN
fcis-10369	235	8	w	w	PROPN
fcis-10369	235	9	,	,	PUNCT
fcis-10369	235	10	wang	wang	PROPN
fcis-10369	235	11	a	a	PROPN
fcis-10369	235	12	,	,	PUNCT
fcis-10369	235	13	et	et	PROPN
fcis-10369	235	14	al	al	PROPN
fcis-10369	235	15	.	.	PUNCT
fcis-10369	235	16	network	network	PROPN
fcis-10369	235	17	intrusion	intrusion	PROPN
fcis-10369	235	18	detection	detection	NOUN
fcis-10369	235	19	combined	combine	VERB
fcis-10369	235	20	hybrid	hybrid	NOUN
fcis-10369	235	21	sampling	sampling	NOUN
fcis-10369	235	22	with	with	ADP
fcis-10369	235	23	deep	deep	ADJ
fcis-10369	235	24	hierarchical	hierarchical	ADJ
fcis-10369	235	25	network[j	network[j	NOUN
fcis-10369	235	26	]	]	PUNCT
fcis-10369	235	27	.	.	PUNCT
fcis-10369	236	1	ieee	ieee	NOUN
fcis-10369	236	2	access	access	NOUN
fcis-10369	236	3	,	,	PUNCT
fcis-10369	236	4	2020	2020	NUM
fcis-10369	236	5	,	,	PUNCT
fcis-10369	236	6	8	8	NUM
fcis-10369	236	7	:	:	SYM
fcis-10369	236	8	32464	32464	NUM
fcis-10369	236	9	-	-	SYM
fcis-10369	236	10	32476	32476	NUM
fcis-10369	236	11	.	.	PUNCT
fcis-10369	237	1	[	[	X
fcis-10369	237	2	15	15	NUM
fcis-10369	237	3	]	]	SYM
fcis-10369	237	4	toupas	toupas	NOUN
fcis-10369	237	5	p	p	NOUN
fcis-10369	237	6	,	,	PUNCT
fcis-10369	237	7	chamou	chamou	PROPN
fcis-10369	237	8	d	d	PROPN
fcis-10369	237	9	,	,	PUNCT
fcis-10369	237	10	giannoutakis	giannoutakis	PROPN
fcis-10369	237	11	k	k	PROPN
fcis-10369	237	12	m	m	PROPN
fcis-10369	237	13	,	,	PUNCT
fcis-10369	237	14	et	et	PROPN
fcis-10369	237	15	al	al	PROPN
fcis-10369	237	16	.	.	PUNCT
fcis-10369	238	1	an	an	DET
fcis-10369	238	2	intrusion	intrusion	NOUN
fcis-10369	238	3	detection	detection	NOUN
fcis-10369	238	4	system	system	NOUN
fcis-10369	238	5	for	for	ADP
fcis-10369	238	6	multi	multi	ADJ
fcis-10369	238	7	-	-	ADJ
fcis-10369	238	8	class	class	ADJ
fcis-10369	238	9	classification	classification	NOUN
fcis-10369	238	10	based	base	VERB
fcis-10369	238	11	on	on	ADP
fcis-10369	238	12	deep	deep	ADJ
fcis-10369	238	13	neural	neural	ADJ
fcis-10369	238	14	networks[c]//2019	networks[c]//2019	NUM
fcis-10369	238	15	18th	18th	ADJ
fcis-10369	238	16	ieee	ieee	NOUN
fcis-10369	238	17	international	international	ADJ
fcis-10369	238	18	conference	conference	NOUN
fcis-10369	238	19	on	on	ADP
fcis-10369	238	20	machine	machine	NOUN
fcis-10369	238	21	learning	learning	NOUN
fcis-10369	238	22	and	and	CCONJ
fcis-10369	238	23	applications	application	NOUN
fcis-10369	238	24	(	(	PUNCT
fcis-10369	238	25	icmla	icmla	NOUN
fcis-10369	238	26	)	)	PUNCT
fcis-10369	238	27	.	.	PUNCT
fcis-10369	239	1	ieee	ieee	PROPN
fcis-10369	239	2	,	,	PUNCT
fcis-10369	239	3	2019	2019	NUM
fcis-10369	239	4	:	:	PUNCT
fcis-10369	239	5	1253	1253	NUM
fcis-10369	239	6	-	-	SYM
fcis-10369	239	7	1258	1258	NUM
fcis-10369	239	8	.	.	PUNCT
fcis-10369	240	1	[	[	X
fcis-10369	240	2	16	16	NUM
fcis-10369	240	3	]	]	X
fcis-10369	240	4	al	al	PROPN
fcis-10369	240	5	-	-	PUNCT
fcis-10369	240	6	turaiki	turaiki	PROPN
fcis-10369	240	7	i	i	PROPN
fcis-10369	240	8	,	,	PUNCT
fcis-10369	240	9	altwaijry	altwaijry	PROPN
fcis-10369	240	10	n.	n.	PROPN
fcis-10369	240	11	a	a	DET
fcis-10369	240	12	convolutional	convolutional	ADJ
fcis-10369	240	13	neural	neural	ADJ
fcis-10369	240	14	network	network	NOUN
fcis-10369	240	15	for	for	ADP
fcis-10369	240	16	improved	improved	ADJ
fcis-10369	240	17	anomaly	anomaly	NOUN
fcis-10369	240	18	-	-	PUNCT
fcis-10369	240	19	based	base	VERB
fcis-10369	240	20	network	network	NOUN
fcis-10369	240	21	intrusion	intrusion	NOUN
fcis-10369	240	22	detection[j	detection[j	PROPN
fcis-10369	240	23	]	]	PUNCT
fcis-10369	240	24	.	.	PUNCT
fcis-10369	241	1	big	big	ADJ
fcis-10369	241	2	data	data	PROPN
fcis-10369	241	3	,	,	PUNCT
fcis-10369	241	4	2021	2021	NUM
fcis-10369	241	5	,	,	PUNCT
fcis-10369	241	6	9(3	9(3	NUM
fcis-10369	241	7	):	):	PUNCT
fcis-10369	241	8	233	233	NUM
fcis-10369	241	9	-	-	SYM
fcis-10369	241	10	252	252	NUM
fcis-10369	241	11	.	.	PUNCT
fcis-10369	242	1	[	[	X
fcis-10369	242	2	17	17	NUM
fcis-10369	242	3	]	]	X
fcis-10369	242	4	shu	shu	NOUN
fcis-10369	242	5	d	d	PROPN
fcis-10369	242	6	,	,	PUNCT
fcis-10369	242	7	leslie	leslie	PROPN
fcis-10369	242	8	n	n	PROPN
fcis-10369	242	9	o	o	NOUN
fcis-10369	242	10	,	,	PUNCT
fcis-10369	242	11	kamhoua	kamhoua	PROPN
fcis-10369	242	12	c	c	PROPN
fcis-10369	242	13	a	a	PROPN
fcis-10369	242	14	,	,	PUNCT
fcis-10369	242	15	et	et	PROPN
fcis-10369	242	16	al	al	PROPN
fcis-10369	242	17	.	.	PROPN
fcis-10369	242	18	generative	generative	ADJ
fcis-10369	242	19	adversarial	adversarial	ADJ
fcis-10369	242	20	attacks	attack	NOUN
fcis-10369	242	21	against	against	ADP
fcis-10369	242	22	intrusion	intrusion	NOUN
fcis-10369	242	23	detection	detection	NOUN
fcis-10369	242	24	systems	system	NOUN
fcis-10369	242	25	using	use	VERB
fcis-10369	242	26	active	active	ADJ
fcis-10369	242	27	learning	learning	NOUN
fcis-10369	243	1	[	[	X
fcis-10369	243	2	c]//	c]//	ADJ
fcis-10369	243	3	proceedings	proceeding	NOUN
fcis-10369	243	4	of	of	ADP
fcis-10369	243	5	the	the	DET
fcis-10369	243	6	2nd	2nd	PROPN
fcis-10369	243	7	acm	acm	NOUN
fcis-10369	243	8	workshop	workshop	NOUN
fcis-10369	243	9	on	on	ADP
fcis-10369	243	10	wireless	wireless	ADJ
fcis-10369	243	11	security	security	NOUN
fcis-10369	243	12	and	and	CCONJ
fcis-10369	243	13	machine	machine	NOUN
fcis-10369	243	14	learning	learning	NOUN
fcis-10369	243	15	.	.	PUNCT
fcis-10369	244	1	2020	2020	NUM
fcis-10369	244	2	:	:	PUNCT
fcis-10369	245	1	1	1	NUM
fcis-10369	245	2	-	-	SYM
fcis-10369	245	3	6	6	NUM
fcis-10369	245	4	.	.	PUNCT
fcis-10369	246	1	[	[	X
fcis-10369	246	2	18	18	NUM
fcis-10369	246	3	]	]	X
fcis-10369	246	4	chen	chen	PROPN
fcis-10369	246	5	j	j	PROPN
fcis-10369	246	6	,	,	PUNCT
fcis-10369	246	7	wu	wu	PROPN
fcis-10369	246	8	d	d	PROPN
fcis-10369	246	9	,	,	PUNCT
fcis-10369	246	10	zhao	zhao	PROPN
fcis-10369	246	11	y	y	PROPN
fcis-10369	246	12	,	,	PUNCT
fcis-10369	246	13	et	et	PROPN
fcis-10369	246	14	al	al	PROPN
fcis-10369	246	15	.	.	PROPN
fcis-10369	246	16	fooling	fooling	PROPN
fcis-10369	246	17	intrusion	intrusion	NOUN
fcis-10369	246	18	detection	detection	NOUN
fcis-10369	246	19	systems	system	NOUN
fcis-10369	246	20	using	use	VERB
fcis-10369	246	21	adversarially	adversarially	ADV
fcis-10369	246	22	autoencoder[j	autoencoder[j	NOUN
fcis-10369	246	23	]	]	PUNCT
fcis-10369	246	24	.	.	PUNCT
fcis-10369	247	1	digital	digital	ADJ
fcis-10369	247	2	communications	communication	NOUN
fcis-10369	247	3	and	and	CCONJ
fcis-10369	247	4	networks	network	NOUN
fcis-10369	247	5	,	,	PUNCT
fcis-10369	247	6	2021	2021	NUM
fcis-10369	247	7	,	,	PUNCT
fcis-10369	247	8	7(3	7(3	NUM
fcis-10369	247	9	):	):	PUNCT
fcis-10369	247	10	453	453	NUM
fcis-10369	247	11	-	-	SYM
fcis-10369	247	12	460	460	NUM
fcis-10369	247	13	.	.	PUNCT
fcis-10369	248	1	[	[	X
fcis-10369	248	2	19	19	NUM
fcis-10369	248	3	]	]	PUNCT
fcis-10369	248	4	yin	yin	PROPN
fcis-10369	248	5	chuanlong	chuanlong	PROPN
fcis-10369	248	6	.	.	PUNCT
fcis-10369	249	1	research	research	NOUN
fcis-10369	249	2	on	on	ADP
fcis-10369	249	3	network	network	NOUN
fcis-10369	249	4	anomaly	anomaly	NOUN
fcis-10369	249	5	detection	detection	NOUN
fcis-10369	249	6	technology	technology	NOUN
fcis-10369	249	7	based	base	VERB
fcis-10369	249	8	on	on	ADP
fcis-10369	249	9	deep	deep	ADJ
fcis-10369	249	10	learning	learning	NOUN
fcis-10369	250	1	[	[	X
fcis-10369	250	2	d	d	X
fcis-10369	250	3	]	]	X
fcis-10369	250	4	.	.	PUNCT
fcis-10369	251	1	information	information	NOUN
fcis-10369	251	2	engineering	engineering	PROPN
fcis-10369	251	3	university	university	PROPN
fcis-10369	251	4	of	of	ADP
fcis-10369	251	5	strategic	strategic	ADJ
fcis-10369	251	6	support	support	NOUN
fcis-10369	251	7	force	force	NOUN
fcis-10369	251	8	,	,	PUNCT
fcis-10369	251	9	2018	2018	NUM
fcis-10369	251	10	.	.	PUNCT
fcis-10369	252	1	[	[	X
fcis-10369	252	2	20	20	NUM
fcis-10369	252	3	]	]	X
fcis-10369	252	4	dong	dong	PROPN
fcis-10369	252	5	weiyu	weiyu	PROPN
fcis-10369	252	6	,	,	PUNCT
fcis-10369	252	7	li	li	PROPN
fcis-10369	252	8	haitao	haitao	PROPN
fcis-10369	252	9	,	,	PUNCT
fcis-10369	252	10	wang	wang	PROPN
fcis-10369	252	11	ruimin	ruimin	PROPN
fcis-10369	252	12	,	,	PUNCT
fcis-10369	252	13	ren	ren	PROPN
fcis-10369	252	14	huajuan	huajuan	PROPN
fcis-10369	252	15	,	,	PUNCT
fcis-10369	252	16	sun	sun	PROPN
fcis-10369	252	17	xuekai	xuekai	PROPN
fcis-10369	252	18	.	.	PUNCT
fcis-10369	253	1	network	network	NOUN
fcis-10369	253	2	traffic	traffic	NOUN
fcis-10369	253	3	anomaly	anomaly	PROPN
fcis-10369	253	4	detection	detection	NOUN
fcis-10369	253	5	model	model	NOUN
fcis-10369	253	6	based	base	VERB
fcis-10369	253	7	on	on	ADP
fcis-10369	253	8	stacked	stack	VERB
fcis-10369	253	9	convolutional	convolutional	ADJ
fcis-10369	253	10	attention[j	attention[j	PROPN
fcis-10369	253	11	]	]	PUNCT
fcis-10369	253	12	.	.	PUNCT
fcis-10369	254	1	computer	computer	NOUN
fcis-10369	254	2	engineering	engineering	NOUN
fcis-10369	254	3	,	,	PUNCT
fcis-10369	254	4	2022	2022	NUM
fcis-10369	254	5	,	,	PUNCT
fcis-10369	254	6	48(09	48(09	NUM
fcis-10369	254	7	):	):	PUNCT
fcis-10369	254	8	12	12	NUM
fcis-10369	254	9	-	-	SYM
fcis-10369	254	10	19	19	NUM
fcis-10369	254	11	.	.	PUNCT