id	sid	tid	token	lemma	pos
cana-4260	1	1	communications	communication	NOUN
cana-4260	1	2	on	on	ADP
cana-4260	1	3	applied	apply	VERB
cana-4260	1	4	nonlinear	nonlinear	ADJ
cana-4260	1	5	analysis	analysis	NOUN
cana-4260	1	6	issn	issn	NOUN
cana-4260	1	7	:	:	PUNCT
cana-4260	1	8	1074	1074	NUM
cana-4260	1	9	-	-	PUNCT
cana-4260	1	10	133x	133x	NUM
cana-4260	1	11	vol	vol	NOUN
cana-4260	1	12	32	32	NUM
cana-4260	1	13	no	no	NOUN
cana-4260	1	14	.	.	PUNCT
cana-4260	2	1	9s	9s	NUM
cana-4260	2	2	(	(	PUNCT
cana-4260	2	3	2025	2025	NUM
cana-4260	2	4	)	)	PUNCT
cana-4260	2	5	1630	1630	NUM
cana-4260	2	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-4260	2	7	advanced	advanced	ADJ
cana-4260	2	8	re	re	VERB
cana-4260	2	9	-	-	VERB
cana-4260	2	10	sampling	sample	VERB
cana-4260	2	11	techniques	technique	NOUN
cana-4260	2	12	for	for	ADP
cana-4260	2	13	multi	multi	ADJ
cana-4260	2	14	-	-	ADJ
cana-4260	2	15	class	class	ADJ
cana-4260	2	16	imbalanced	imbalanced	ADJ
cana-4260	2	17	classification	classification	NOUN
cana-4260	2	18	k.v.chandra	k.v.chandra	NOUN
cana-4260	2	19	sekhar1	sekhar1	PROPN
cana-4260	2	20	,	,	PUNCT
cana-4260	2	21	balaka	balaka	PROPN
cana-4260	2	22	ramjee2	ramjee2	PROPN
cana-4260	2	23	,	,	PUNCT
cana-4260	2	24	landa	landa	PROPN
cana-4260	2	25	naresh3	naresh3	PROPN
cana-4260	2	26	,	,	PUNCT
cana-4260	2	27	pasala	pasala	PROPN
cana-4260	2	28	mahesh4	mahesh4	NOUN
cana-4260	2	29	,	,	PUNCT
cana-4260	2	30	kinthada	kinthada	PROPN
cana-4260	2	31	jayaramu5	jayaramu5	NOUN
cana-4260	2	32	1assistant	1assistant	PROPN
cana-4260	2	33	.	.	PUNCT
cana-4260	2	34	professor	professor	NOUN
cana-4260	2	35	,	,	PUNCT
cana-4260	2	36	department	department	NOUN
cana-4260	2	37	of	of	ADP
cana-4260	2	38	cse(ai&ml	cse(ai&ml	PROPN
cana-4260	2	39	)	)	PUNCT
cana-4260	2	40	,	,	PUNCT
cana-4260	2	41	aditya	aditya	PROPN
cana-4260	2	42	institute	institute	PROPN
cana-4260	2	43	of	of	ADP
cana-4260	2	44	technology	technology	NOUN
cana-4260	2	45	and	and	CCONJ
cana-4260	2	46	management	management	NOUN
cana-4260	2	47	,	,	PUNCT
cana-4260	2	48	tekkali532201	tekkali532201	PROPN
cana-4260	2	49	,	,	PUNCT
cana-4260	2	50	india	india	PROPN
cana-4260	2	51	.	.	PUNCT
cana-4260	3	1	2,3,4,5ug	2,3,4,5ug	NUM
cana-4260	3	2	students	student	NOUN
cana-4260	3	3	,	,	PUNCT
cana-4260	3	4	department	department	NOUN
cana-4260	3	5	of	of	ADP
cana-4260	3	6	computer	computer	NOUN
cana-4260	3	7	science	science	NOUN
cana-4260	3	8	and	and	CCONJ
cana-4260	3	9	engineering	engineering	NOUN
cana-4260	3	10	(	(	PUNCT
cana-4260	3	11	ai&ml	ai&ml	NOUN
cana-4260	3	12	)	)	PUNCT
cana-4260	3	13	,	,	PUNCT
cana-4260	3	14	aditya	aditya	PROPN
cana-4260	3	15	institute	institute	PROPN
cana-4260	3	16	of	of	ADP
cana-4260	3	17	technology	technology	NOUN
cana-4260	3	18	and	and	CCONJ
cana-4260	3	19	management	management	NOUN
cana-4260	3	20	,	,	PUNCT
cana-4260	3	21	tekkali-532201	tekkali-532201	ADJ
cana-4260	3	22	,	,	PUNCT
cana-4260	3	23	india	india	PROPN
cana-4260	3	24	.	.	PUNCT
cana-4260	4	1	article	article	PROPN
cana-4260	4	2	history	history	NOUN
cana-4260	4	3	:	:	PUNCT
cana-4260	4	4	received	receive	VERB
cana-4260	4	5	:	:	PUNCT
cana-4260	4	6	12	12	NUM
cana-4260	4	7	-	-	SYM
cana-4260	4	8	01	01	NUM
cana-4260	4	9	-	-	PUNCT
cana-4260	4	10	2025	2025	NUM
cana-4260	4	11	revised	revise	VERB
cana-4260	4	12	:	:	PUNCT
cana-4260	4	13	15	15	NUM
cana-4260	4	14	-	-	NUM
cana-4260	4	15	02	02	NUM
cana-4260	4	16	-	-	PUNCT
cana-4260	4	17	2025	2025	NUM
cana-4260	4	18	accepted	accept	VERB
cana-4260	4	19	:	:	PUNCT
cana-4260	4	20	01	01	NUM
cana-4260	4	21	-	-	SYM
cana-4260	4	22	03	03	NUM
cana-4260	4	23	-	-	PUNCT
cana-4260	4	24	2025	2025	NUM
cana-4260	4	25	abstract	abstract	NOUN
cana-4260	4	26	:	:	PUNCT
cana-4260	4	27	imbalanced	imbalanced	ADJ
cana-4260	4	28	classification	classification	NOUN
cana-4260	4	29	is	be	AUX
cana-4260	4	30	a	a	DET
cana-4260	4	31	common	common	ADJ
cana-4260	4	32	problem	problem	NOUN
cana-4260	4	33	in	in	ADP
cana-4260	4	34	machine	machine	NOUN
cana-4260	4	35	learning	learning	NOUN
cana-4260	4	36	,	,	PUNCT
cana-4260	4	37	where	where	SCONJ
cana-4260	4	38	one	one	NUM
cana-4260	4	39	class	class	NOUN
cana-4260	4	40	significantly	significantly	ADV
cana-4260	4	41	outnumbers	outnumber	VERB
cana-4260	4	42	the	the	DET
cana-4260	4	43	others	other	NOUN
cana-4260	4	44	.	.	PUNCT
cana-4260	5	1	this	this	DET
cana-4260	5	2	imbalance	imbalance	NOUN
cana-4260	5	3	leads	lead	VERB
cana-4260	5	4	to	to	ADP
cana-4260	5	5	biased	biased	ADJ
cana-4260	5	6	model	model	NOUN
cana-4260	5	7	performance	performance	NOUN
cana-4260	5	8	,	,	PUNCT
cana-4260	5	9	where	where	SCONJ
cana-4260	5	10	the	the	DET
cana-4260	5	11	classifier	classifier	NOUN
cana-4260	5	12	favors	favor	VERB
cana-4260	5	13	the	the	DET
cana-4260	5	14	majority	majority	NOUN
cana-4260	5	15	class	class	NOUN
cana-4260	5	16	,	,	PUNCT
cana-4260	5	17	resulting	result	VERB
cana-4260	5	18	in	in	ADP
cana-4260	5	19	poor	poor	ADJ
cana-4260	5	20	detection	detection	NOUN
cana-4260	5	21	of	of	ADP
cana-4260	5	22	the	the	DET
cana-4260	5	23	minority	minority	NOUN
cana-4260	5	24	class	class	NOUN
cana-4260	5	25	.	.	PUNCT
cana-4260	6	1	traditional	traditional	ADJ
cana-4260	6	2	machine	machine	NOUN
cana-4260	6	3	learning	learn	VERB
cana-4260	6	4	algorithms	algorithm	NOUN
cana-4260	6	5	assume	assume	VERB
cana-4260	6	6	a	a	DET
cana-4260	6	7	balanced	balanced	ADJ
cana-4260	6	8	distribution	distribution	NOUN
cana-4260	6	9	,	,	PUNCT
cana-4260	6	10	making	make	VERB
cana-4260	6	11	them	they	PRON
cana-4260	6	12	ineffective	ineffective	ADJ
cana-4260	6	13	in	in	ADP
cana-4260	6	14	such	such	ADJ
cana-4260	6	15	scenarios	scenario	NOUN
cana-4260	6	16	.	.	PUNCT
cana-4260	7	1	various	various	ADJ
cana-4260	7	2	techniques	technique	NOUN
cana-4260	7	3	,	,	PUNCT
cana-4260	7	4	including	include	VERB
cana-4260	7	5	resampling	resample	VERB
cana-4260	7	6	methods	method	NOUN
cana-4260	7	7	(	(	PUNCT
cana-4260	7	8	such	such	ADJ
cana-4260	7	9	as	as	ADP
cana-4260	7	10	oversampling	oversample	VERB
cana-4260	7	11	and	and	CCONJ
cana-4260	7	12	undersampling	undersampling	ADJ
cana-4260	7	13	)	)	PUNCT
cana-4260	7	14	,	,	PUNCT
cana-4260	7	15	cost	cost	NOUN
cana-4260	7	16	-	-	PUNCT
cana-4260	7	17	sensitive	sensitive	ADJ
cana-4260	7	18	learning	learning	NOUN
cana-4260	7	19	,	,	PUNCT
cana-4260	7	20	and	and	CCONJ
cana-4260	7	21	synthetic	synthetic	ADJ
cana-4260	7	22	data	data	NOUN
cana-4260	7	23	generation	generation	NOUN
cana-4260	7	24	,	,	PUNCT
cana-4260	7	25	have	have	AUX
cana-4260	7	26	been	be	AUX
cana-4260	7	27	proposed	propose	VERB
cana-4260	7	28	to	to	PART
cana-4260	7	29	address	address	VERB
cana-4260	7	30	this	this	DET
cana-4260	7	31	challenge	challenge	NOUN
cana-4260	7	32	.	.	PUNCT
cana-4260	8	1	effective	effective	ADJ
cana-4260	8	2	handling	handling	NOUN
cana-4260	8	3	of	of	ADP
cana-4260	8	4	imbalanced	imbalanced	ADJ
cana-4260	8	5	data	datum	NOUN
cana-4260	8	6	is	be	AUX
cana-4260	8	7	crucial	crucial	ADJ
cana-4260	8	8	in	in	ADP
cana-4260	8	9	applications	application	NOUN
cana-4260	8	10	like	like	ADP
cana-4260	8	11	fraud	fraud	NOUN
cana-4260	8	12	detection	detection	NOUN
cana-4260	8	13	,	,	PUNCT
cana-4260	8	14	medical	medical	ADJ
cana-4260	8	15	diagnosis	diagnosis	NOUN
cana-4260	8	16	,	,	PUNCT
cana-4260	8	17	and	and	CCONJ
cana-4260	8	18	anomaly	anomaly	NOUN
cana-4260	8	19	detection	detection	NOUN
cana-4260	8	20	,	,	PUNCT
cana-4260	8	21	where	where	SCONJ
cana-4260	8	22	minority	minority	NOUN
cana-4260	8	23	class	class	NOUN
cana-4260	8	24	predictions	prediction	NOUN
cana-4260	8	25	hold	hold	VERB
cana-4260	8	26	high	high	ADJ
cana-4260	8	27	significance	significance	NOUN
cana-4260	8	28	.	.	PUNCT
cana-4260	9	1	this	this	DET
cana-4260	9	2	study	study	NOUN
cana-4260	9	3	explores	explore	VERB
cana-4260	9	4	different	different	ADJ
cana-4260	9	5	approaches	approach	NOUN
cana-4260	9	6	to	to	PART
cana-4260	9	7	mitigate	mitigate	VERB
cana-4260	9	8	class	class	NOUN
cana-4260	9	9	imbalance	imbalance	NOUN
cana-4260	9	10	and	and	CCONJ
cana-4260	9	11	improve	improve	VERB
cana-4260	9	12	classification	classification	NOUN
cana-4260	9	13	performance	performance	NOUN
cana-4260	9	14	,	,	PUNCT
cana-4260	9	15	ensuring	ensure	VERB
cana-4260	9	16	better	well	ADJ
cana-4260	9	17	generalization	generalization	NOUN
cana-4260	9	18	and	and	CCONJ
cana-4260	9	19	robustness	robustness	NOUN
cana-4260	9	20	in	in	ADP
cana-4260	9	21	real	real	ADJ
cana-4260	9	22	-	-	PUNCT
cana-4260	9	23	world	world	NOUN
cana-4260	9	24	scenarios	scenario	NOUN
cana-4260	9	25	.	.	PUNCT
cana-4260	10	1	introduction	introduction	NOUN
cana-4260	10	2	:	:	PUNCT
cana-4260	10	3	model	model	NOUN
cana-4260	10	4	predictions	prediction	NOUN
cana-4260	10	5	are	be	AUX
cana-4260	10	6	skewed	skew	VERB
cana-4260	10	7	by	by	ADP
cana-4260	10	8	class	class	NOUN
cana-4260	10	9	imbalances	imbalance	NOUN
cana-4260	10	10	,	,	PUNCT
cana-4260	10	11	rendering	rendering	NOUN
cana-4260	10	12	accuracy	accuracy	NOUN
cana-4260	10	13	metrics	metric	NOUN
cana-4260	10	14	less	less	ADV
cana-4260	10	15	meaningful	meaningful	ADJ
cana-4260	10	16	as	as	SCONJ
cana-4260	10	17	is	be	AUX
cana-4260	10	18	often	often	ADV
cana-4260	10	19	the	the	DET
cana-4260	10	20	case	case	NOUN
cana-4260	10	21	in	in	ADP
cana-4260	10	22	healthcare	healthcare	PROPN
cana-4260	10	23	and	and	CCONJ
cana-4260	10	24	fraud	fraud	NOUN
cana-4260	10	25	detection	detection	NOUN
cana-4260	10	26	.	.	PUNCT
cana-4260	11	1	smote	smote	VERB
cana-4260	11	2	and	and	CCONJ
cana-4260	11	3	its	its	PRON
cana-4260	11	4	derivatives	derivative	NOUN
cana-4260	11	5	are	be	AUX
cana-4260	11	6	also	also	ADV
cana-4260	11	7	an	an	DET
cana-4260	11	8	example	example	NOUN
cana-4260	11	9	of	of	ADP
cana-4260	11	10	resampling	resample	VERB
cana-4260	11	11	techniques	technique	NOUN
cana-4260	11	12	which	which	PRON
cana-4260	11	13	creates	create	VERB
cana-4260	11	14	synthetic	synthetic	ADJ
cana-4260	11	15	data	datum	NOUN
cana-4260	11	16	to	to	PART
cana-4260	11	17	balance	balance	VERB
cana-4260	11	18	the	the	DET
cana-4260	11	19	classes	class	NOUN
cana-4260	11	20	for	for	ADP
cana-4260	11	21	better	well	ADJ
cana-4260	11	22	learning	learning	NOUN
cana-4260	11	23	.	.	PUNCT
cana-4260	12	1	such	such	ADJ
cana-4260	12	2	as	as	ADP
cana-4260	12	3	borderline	borderline	NOUN
cana-4260	12	4	-	-	PUNCT
cana-4260	12	5	smote	smote	ADJ
cana-4260	12	6	,	,	PUNCT
cana-4260	12	7	adasyn	adasyn	PROPN
cana-4260	12	8	,	,	PUNCT
cana-4260	12	9	smoteenn	smoteenn	NOUN
cana-4260	12	10	or	or	CCONJ
cana-4260	12	11	smotetomek	smotetomek	PROPN
cana-4260	12	12	helped	help	VERB
cana-4260	12	13	to	to	PART
cana-4260	12	14	improve	improve	VERB
cana-4260	12	15	the	the	DET
cana-4260	12	16	decision	decision	NOUN
cana-4260	12	17	boundaries	boundary	NOUN
cana-4260	12	18	and	and	CCONJ
cana-4260	12	19	noise	noise	NOUN
cana-4260	12	20	reduction	reduction	NOUN
cana-4260	12	21	.	.	PUNCT
cana-4260	13	1	these	these	DET
cana-4260	13	2	techniques	technique	NOUN
cana-4260	13	3	facilitate	facilitate	VERB
cana-4260	13	4	creation	creation	NOUN
cana-4260	13	5	of	of	ADP
cana-4260	13	6	better	well	ADJ
cana-4260	13	7	models	model	NOUN
cana-4260	13	8	by	by	ADP
cana-4260	13	9	addressing	address	VERB
cana-4260	13	10	the	the	DET
cana-4260	13	11	issue	issue	NOUN
cana-4260	13	12	where	where	SCONJ
cana-4260	13	13	minority	minority	NOUN
cana-4260	13	14	class	class	NOUN
cana-4260	13	15	is	be	AUX
cana-4260	13	16	n't	not	PART
cana-4260	13	17	represented	represent	VERB
cana-4260	13	18	in	in	ADP
cana-4260	13	19	feature	feature	NOUN
cana-4260	13	20	space	space	NOUN
cana-4260	13	21	fairly	fairly	ADV
cana-4260	13	22	.	.	PUNCT
cana-4260	14	1	methodology	methodology	NOUN
cana-4260	14	2	:	:	PUNCT
cana-4260	14	3	the	the	DET
cana-4260	14	4	gmm	gmm	NOUN
cana-4260	14	5	-	-	PUNCT
cana-4260	14	6	smote	smote	ADJ
cana-4260	14	7	method	method	NOUN
cana-4260	14	8	addresses	address	NOUN
cana-4260	14	9	imbalanced	imbalanced	ADJ
cana-4260	14	10	datasets	dataset	NOUN
cana-4260	14	11	by	by	ADP
cana-4260	14	12	utilizing	utilize	VERB
cana-4260	14	13	gaussian	gaussian	ADJ
cana-4260	14	14	mixture	mixture	NOUN
cana-4260	14	15	model	model	NOUN
cana-4260	14	16	(	(	PUNCT
cana-4260	14	17	gmm	gmm	NOUN
cana-4260	14	18	)	)	PUNCT
cana-4260	14	19	for	for	ADP
cana-4260	14	20	clustering	cluster	VERB
cana-4260	14	21	and	and	CCONJ
cana-4260	14	22	applying	apply	VERB
cana-4260	14	23	smote	smote	NOUN
cana-4260	14	24	to	to	ADP
cana-4260	14	25	oversample	oversample	ADJ
cana-4260	14	26	minority	minority	NOUN
cana-4260	14	27	data	datum	NOUN
cana-4260	14	28	in	in	ADP
cana-4260	14	29	high	high	ADJ
cana-4260	14	30	-	-	PUNCT
cana-4260	14	31	density	density	NOUN
cana-4260	14	32	areas	area	NOUN
cana-4260	14	33	.	.	PUNCT
cana-4260	15	1	this	this	DET
cana-4260	15	2	approach	approach	NOUN
cana-4260	15	3	involves	involve	VERB
cana-4260	15	4	clustering	clustering	ADJ
cana-4260	15	5	data	datum	NOUN
cana-4260	15	6	,	,	PUNCT
cana-4260	15	7	selecting	select	VERB
cana-4260	15	8	clusters	cluster	NOUN
cana-4260	15	9	with	with	ADP
cana-4260	15	10	significant	significant	ADJ
cana-4260	15	11	minority	minority	NOUN
cana-4260	15	12	presence	presence	NOUN
cana-4260	15	13	,	,	PUNCT
cana-4260	15	14	and	and	CCONJ
cana-4260	15	15	generating	generate	VERB
cana-4260	15	16	synthetic	synthetic	ADJ
cana-4260	15	17	samples	sample	NOUN
cana-4260	15	18	to	to	PART
cana-4260	15	19	ensure	ensure	VERB
cana-4260	15	20	better	well	ADJ
cana-4260	15	21	balance	balance	NOUN
cana-4260	15	22	.	.	PUNCT
cana-4260	16	1	gmm	gmm	PROPN
cana-4260	16	2	enhances	enhance	VERB
cana-4260	16	3	clustering	cluster	VERB
cana-4260	16	4	by	by	ADP
cana-4260	16	5	assigning	assign	VERB
cana-4260	16	6	probabilities	probability	NOUN
cana-4260	16	7	to	to	ADP
cana-4260	16	8	data	datum	NOUN
cana-4260	16	9	points	point	NOUN
cana-4260	16	10	,	,	PUNCT
cana-4260	16	11	while	while	SCONJ
cana-4260	16	12	smote	smote	ADJ
cana-4260	16	13	focuses	focus	VERB
cana-4260	16	14	on	on	ADP
cana-4260	16	15	producing	produce	VERB
cana-4260	16	16	samples	sample	NOUN
cana-4260	16	17	in	in	ADP
cana-4260	16	18	less	less	ADJ
cana-4260	16	19	populated	populated	ADJ
cana-4260	16	20	regions	region	NOUN
cana-4260	16	21	,	,	PUNCT
cana-4260	16	22	effectively	effectively	ADV
cana-4260	16	23	reducing	reduce	VERB
cana-4260	16	24	noise	noise	NOUN
cana-4260	16	25	and	and	CCONJ
cana-4260	16	26	improving	improve	VERB
cana-4260	16	27	class	class	NOUN
cana-4260	16	28	representation	representation	NOUN
cana-4260	16	29	and	and	CCONJ
cana-4260	16	30	model	model	NOUN
cana-4260	16	31	performance	performance	NOUN
cana-4260	16	32	in	in	ADP
cana-4260	16	33	imbalanced	imbalanced	ADJ
cana-4260	16	34	situations	situation	NOUN
cana-4260	16	35	.	.	PUNCT
cana-4260	17	1	results	result	NOUN
cana-4260	17	2	:	:	PUNCT
cana-4260	17	3	the	the	DET
cana-4260	17	4	study	study	NOUN
cana-4260	17	5	evaluates	evaluate	VERB
cana-4260	17	6	gmm	gmm	NOUN
cana-4260	17	7	-	-	PUNCT
cana-4260	17	8	smote	smote	ADJ
cana-4260	17	9	against	against	ADP
cana-4260	17	10	various	various	ADJ
cana-4260	17	11	oversampling	oversampling	ADJ
cana-4260	17	12	techniques	technique	NOUN
cana-4260	17	13	,	,	PUNCT
cana-4260	17	14	including	include	VERB
cana-4260	17	15	kmeans	kmean	NOUN
cana-4260	17	16	-	-	PUNCT
cana-4260	17	17	smote	smote	ADJ
cana-4260	17	18	,	,	PUNCT
cana-4260	17	19	kmeans	kmean	NOUN
cana-4260	17	20	-	-	PUNCT
cana-4260	17	21	adasyn	adasyn	NOUN
cana-4260	17	22	,	,	PUNCT
cana-4260	17	23	and	and	CCONJ
cana-4260	17	24	gmm	gmm	NOUN
cana-4260	17	25	-	-	PUNCT
cana-4260	17	26	adasyn	adasyn	PROPN
cana-4260	17	27	,	,	PUNCT
cana-4260	17	28	using	use	VERB
cana-4260	17	29	datasets	dataset	NOUN
cana-4260	17	30	such	such	ADJ
cana-4260	17	31	as	as	ADP
cana-4260	17	32	breast	breast	NOUN
cana-4260	17	33	cancer	cancer	NOUN
cana-4260	17	34	,	,	PUNCT
cana-4260	17	35	crx	crx	PROPN
cana-4260	17	36	,	,	PUNCT
cana-4260	17	37	and	and	CCONJ
cana-4260	17	38	churn	churn	VERB
cana-4260	17	39	bigml	bigml	PROPN
cana-4260	17	40	.	.	PUNCT
cana-4260	18	1	performance	performance	NOUN
cana-4260	18	2	metrics	metric	NOUN
cana-4260	18	3	include	include	VERB
cana-4260	18	4	accuracy	accuracy	NOUN
cana-4260	18	5	,	,	PUNCT
cana-4260	18	6	auc	auc	NOUN
cana-4260	18	7	-	-	PUNCT
cana-4260	18	8	roc	roc	NOUN
cana-4260	18	9	score	score	NOUN
cana-4260	18	10	,	,	PUNCT
cana-4260	18	11	and	and	CCONJ
cana-4260	18	12	computational	computational	ADJ
cana-4260	18	13	efficiency	efficiency	NOUN
cana-4260	18	14	across	across	ADP
cana-4260	18	15	classifiers	classifier	NOUN
cana-4260	18	16	like	like	ADP
cana-4260	18	17	random	random	ADJ
cana-4260	18	18	forest	forest	NOUN
cana-4260	18	19	,	,	PUNCT
cana-4260	18	20	svm	svm	ADJ
cana-4260	18	21	,	,	PUNCT
cana-4260	18	22	logistic	logistic	ADJ
cana-4260	18	23	regression	regression	NOUN
cana-4260	18	24	,	,	PUNCT
cana-4260	18	25	and	and	CCONJ
cana-4260	18	26	neural	neural	ADJ
cana-4260	18	27	networks	network	NOUN
cana-4260	18	28	.	.	PUNCT
cana-4260	19	1	results	result	NOUN
cana-4260	19	2	demonstrate	demonstrate	VERB
cana-4260	19	3	that	that	SCONJ
cana-4260	19	4	gmm	gmm	NOUN
cana-4260	19	5	-	-	PUNCT
cana-4260	19	6	smote	smote	ADJ
cana-4260	19	7	enhances	enhance	VERB
cana-4260	19	8	classification	classification	NOUN
cana-4260	19	9	through	through	ADP
cana-4260	19	10	balanced	balanced	ADJ
cana-4260	19	11	decision	decision	NOUN
cana-4260	19	12	boundaries	boundary	NOUN
cana-4260	19	13	and	and	CCONJ
cana-4260	19	14	shows	show	VERB
cana-4260	19	15	efficiency	efficiency	NOUN
cana-4260	19	16	in	in	ADP
cana-4260	19	17	training	training	NOUN
cana-4260	19	18	time	time	NOUN
cana-4260	19	19	,	,	PUNCT
cana-4260	19	20	making	make	VERB
cana-4260	19	21	it	it	PRON
cana-4260	19	22	advantageous	advantageous	ADJ
cana-4260	19	23	for	for	SCONJ
cana-4260	19	24	communications	communication	NOUN
cana-4260	19	25	on	on	ADP
cana-4260	19	26	applied	apply	VERB
cana-4260	19	27	nonlinear	nonlinear	ADJ
cana-4260	19	28	analysis	analysis	NOUN
cana-4260	19	29	issn	issn	NOUN
cana-4260	19	30	:	:	PUNCT
cana-4260	19	31	1074	1074	NUM
cana-4260	19	32	-	-	PUNCT
cana-4260	19	33	133x	133x	NUM
cana-4260	19	34	vol	vol	NOUN
cana-4260	19	35	32	32	NUM
cana-4260	19	36	no	no	NOUN
cana-4260	19	37	.	.	PUNCT
cana-4260	20	1	9s	9s	NUM
cana-4260	20	2	(	(	PUNCT
cana-4260	20	3	2025	2025	NUM
cana-4260	20	4	)	)	PUNCT
cana-4260	20	5	1631	1631	NUM
cana-4260	21	1	https://internationalpubls.com	https://internationalpubls.com	X
cana-4260	21	2	managing	manage	VERB
cana-4260	21	3	imbalanced	imbalanced	ADJ
cana-4260	21	4	datasets	dataset	NOUN
cana-4260	21	5	.	.	PUNCT
cana-4260	22	1	conclusions	conclusion	NOUN
cana-4260	22	2	:	:	PUNCT
cana-4260	22	3	the	the	DET
cana-4260	22	4	study	study	NOUN
cana-4260	22	5	assesses	assess	VERB
cana-4260	22	6	the	the	DET
cana-4260	22	7	effectiveness	effectiveness	NOUN
cana-4260	22	8	of	of	ADP
cana-4260	22	9	gmm	gmm	NOUN
cana-4260	22	10	-	-	PUNCT
cana-4260	22	11	smote	smote	ADJ
cana-4260	22	12	in	in	ADP
cana-4260	22	13	enhancing	enhance	VERB
cana-4260	22	14	minority	minority	NOUN
cana-4260	22	15	class	class	NOUN
cana-4260	22	16	representation	representation	NOUN
cana-4260	22	17	and	and	CCONJ
cana-4260	22	18	maintaining	maintain	VERB
cana-4260	22	19	balanced	balanced	ADJ
cana-4260	22	20	decision	decision	NOUN
cana-4260	22	21	boundaries	boundary	NOUN
cana-4260	22	22	compared	compare	VERB
cana-4260	22	23	to	to	ADP
cana-4260	22	24	traditional	traditional	ADJ
cana-4260	22	25	oversampling	oversampling	ADJ
cana-4260	22	26	methods	method	NOUN
cana-4260	22	27	like	like	ADP
cana-4260	22	28	smote	smote	NOUN
cana-4260	22	29	and	and	CCONJ
cana-4260	22	30	adasyn	adasyn	PROPN
cana-4260	22	31	.	.	PUNCT
cana-4260	23	1	gmm	gmm	PROPN
cana-4260	23	2	-	-	PUNCT
cana-4260	23	3	smote	smote	ADJ
cana-4260	23	4	generates	generate	VERB
cana-4260	23	5	more	more	ADV
cana-4260	23	6	meaningful	meaningful	ADJ
cana-4260	23	7	synthetic	synthetic	ADJ
cana-4260	23	8	samples	sample	NOUN
cana-4260	23	9	and	and	CCONJ
cana-4260	23	10	mitigates	mitigate	NOUN
cana-4260	23	11	overfitting	overfitte	VERB
cana-4260	23	12	.	.	PUNCT
cana-4260	24	1	future	future	ADJ
cana-4260	24	2	research	research	NOUN
cana-4260	24	3	will	will	AUX
cana-4260	24	4	focus	focus	VERB
cana-4260	24	5	on	on	ADP
cana-4260	24	6	adaptive	adaptive	ADJ
cana-4260	24	7	parameter	parameter	NOUN
cana-4260	24	8	tuning	tuning	NOUN
cana-4260	24	9	,	,	PUNCT
cana-4260	24	10	integration	integration	NOUN
cana-4260	24	11	with	with	ADP
cana-4260	24	12	deep	deep	ADJ
cana-4260	24	13	learning	learning	NOUN
cana-4260	24	14	,	,	PUNCT
cana-4260	24	15	and	and	CCONJ
cana-4260	24	16	real	real	ADJ
cana-4260	24	17	-	-	PUNCT
cana-4260	24	18	time	time	NOUN
cana-4260	24	19	applications	application	NOUN
cana-4260	24	20	,	,	PUNCT
cana-4260	24	21	with	with	ADP
cana-4260	24	22	additional	additional	ADJ
cana-4260	24	23	exploration	exploration	NOUN
cana-4260	24	24	into	into	ADP
cana-4260	24	25	its	its	PRON
cana-4260	24	26	effects	effect	NOUN
cana-4260	24	27	on	on	ADP
cana-4260	24	28	multi	multi	ADJ
cana-4260	24	29	-	-	ADJ
cana-4260	24	30	class	class	ADJ
cana-4260	24	31	imbalance	imbalance	NOUN
cana-4260	24	32	and	and	CCONJ
cana-4260	24	33	computational	computational	ADJ
cana-4260	24	34	efficiency	efficiency	NOUN
cana-4260	24	35	.	.	PUNCT
cana-4260	25	1	overall	overall	ADV
cana-4260	25	2	,	,	PUNCT
cana-4260	25	3	gmm	gmm	NOUN
cana-4260	25	4	-	-	PUNCT
cana-4260	25	5	smote	smote	PROPN
cana-4260	25	6	stands	stand	VERB
cana-4260	25	7	out	out	ADP
cana-4260	25	8	as	as	ADP
cana-4260	25	9	a	a	DET
cana-4260	25	10	valuable	valuable	ADJ
cana-4260	25	11	resampling	resampling	NOUN
cana-4260	25	12	method	method	NOUN
cana-4260	25	13	for	for	ADP
cana-4260	25	14	improving	improve	VERB
cana-4260	25	15	classification	classification	NOUN
cana-4260	25	16	performance	performance	NOUN
cana-4260	25	17	in	in	ADP
cana-4260	25	18	imbalanced	imbalanced	ADJ
cana-4260	25	19	datasets	dataset	NOUN
cana-4260	25	20	.	.	PUNCT
cana-4260	26	1	keywords	keyword	NOUN
cana-4260	26	2	:	:	PUNCT
cana-4260	26	3	kmeanssmote	kmeanssmote	VERB
cana-4260	26	4	,	,	PUNCT
cana-4260	26	5	kmeanssmote	kmeanssmote	VERB
cana-4260	26	6	with	with	ADP
cana-4260	26	7	gmm	gmm	PROPN
cana-4260	26	8	.	.	PUNCT
cana-4260	27	1	1	1	X
cana-4260	27	2	.	.	X
cana-4260	27	3	introduction	introduction	NOUN
cana-4260	27	4	the	the	DET
cana-4260	27	5	dataset	dataset	NOUN
cana-4260	27	6	suffers	suffer	VERB
cana-4260	27	7	from	from	ADP
cana-4260	27	8	the	the	DET
cana-4260	27	9	problem	problem	NOUN
cana-4260	27	10	of	of	ADP
cana-4260	27	11	class	class	NOUN
cana-4260	27	12	imbalance	imbalance	NOUN
cana-4260	27	13	when	when	SCONJ
cana-4260	27	14	one	one	NUM
cana-4260	27	15	class	class	NOUN
cana-4260	27	16	in	in	ADP
cana-4260	27	17	a	a	DET
cana-4260	27	18	dataset	dataset	NOUN
cana-4260	27	19	has	have	VERB
cana-4260	27	20	significantly	significantly	ADV
cana-4260	27	21	more	more	ADJ
cana-4260	27	22	samples	sample	NOUN
cana-4260	27	23	than	than	ADP
cana-4260	27	24	others	other	NOUN
cana-4260	27	25	,	,	PUNCT
cana-4260	27	26	leading	lead	VERB
cana-4260	27	27	to	to	ADP
cana-4260	27	28	biased	biased	ADJ
cana-4260	27	29	model	model	NOUN
cana-4260	27	30	predictions	prediction	NOUN
cana-4260	27	31	and	and	CCONJ
cana-4260	27	32	favoring	favor	VERB
cana-4260	27	33	the	the	DET
cana-4260	27	34	majority	majority	NOUN
cana-4260	27	35	class	class	NOUN
cana-4260	27	36	in	in	ADP
cana-4260	27	37	general	general	ADJ
cana-4260	27	38	,	,	PUNCT
cana-4260	27	39	mostly	mostly	ADV
cana-4260	27	40	neglecting	neglect	VERB
cana-4260	27	41	the	the	DET
cana-4260	27	42	minority	minority	NOUN
cana-4260	27	43	.	.	PUNCT
cana-4260	28	1	for	for	ADP
cana-4260	28	2	example	example	NOUN
cana-4260	28	3	,	,	PUNCT
cana-4260	28	4	while	while	SCONJ
cana-4260	28	5	using	use	VERB
cana-4260	28	6	accuracy	accuracy	NOUN
cana-4260	28	7	metrics	metric	NOUN
cana-4260	28	8	for	for	ADP
cana-4260	28	9	evaluation	evaluation	NOUN
cana-4260	28	10	can	can	AUX
cana-4260	28	11	sometimes	sometimes	ADV
cana-4260	28	12	be	be	AUX
cana-4260	28	13	very	very	ADV
cana-4260	28	14	misleading	misleading	ADJ
cana-4260	28	15	,	,	PUNCT
cana-4260	28	16	accuracy	accuracy	NOUN
cana-4260	28	17	measures	measure	NOUN
cana-4260	28	18	may	may	AUX
cana-4260	28	19	tell	tell	VERB
cana-4260	28	20	the	the	DET
cana-4260	28	21	story	story	NOUN
cana-4260	28	22	differently	differently	ADV
cana-4260	28	23	when	when	SCONJ
cana-4260	28	24	the	the	DET
cana-4260	28	25	minority	minority	NOUN
cana-4260	28	26	is	be	AUX
cana-4260	28	27	in	in	ADP
cana-4260	28	28	consideration	consideration	NOUN
cana-4260	28	29	.	.	PUNCT
cana-4260	29	1	real	real	ADJ
cana-4260	29	2	-	-	PUNCT
cana-4260	29	3	world	world	NOUN
cana-4260	29	4	datasets	dataset	NOUN
cana-4260	29	5	suffer	suffer	VERB
cana-4260	29	6	from	from	ADP
cana-4260	29	7	the	the	DET
cana-4260	29	8	problem	problem	NOUN
cana-4260	29	9	of	of	ADP
cana-4260	29	10	class	class	NOUN
cana-4260	29	11	imbalance	imbalance	NOUN
cana-4260	29	12	in	in	ADP
cana-4260	29	13	almost	almost	ADV
cana-4260	29	14	every	every	PRON
cana-4260	29	15	domain	domain	NOUN
cana-4260	29	16	,	,	PUNCT
cana-4260	29	17	such	such	ADJ
cana-4260	29	18	as	as	ADP
cana-4260	29	19	healthcare	healthcare	NOUN
cana-4260	29	20	and	and	CCONJ
cana-4260	29	21	fraud	fraud	NOUN
cana-4260	29	22	detection	detection	NOUN
cana-4260	29	23	.	.	PUNCT
cana-4260	30	1	there	there	ADV
cana-4260	30	2	,	,	PUNCT
cana-4260	30	3	the	the	DET
cana-4260	30	4	minority	minority	NOUN
cana-4260	30	5	class	class	NOUN
cana-4260	30	6	,	,	PUNCT
cana-4260	30	7	such	such	ADJ
cana-4260	30	8	as	as	ADP
cana-4260	30	9	fraud	fraud	NOUN
cana-4260	30	10	cases	case	NOUN
cana-4260	30	11	or	or	CCONJ
cana-4260	30	12	disease	disease	NOUN
cana-4260	30	13	-	-	PUNCT
cana-4260	30	14	positive	positive	ADJ
cana-4260	30	15	cases	case	NOUN
cana-4260	30	16	,	,	PUNCT
cana-4260	30	17	has	have	VERB
cana-4260	30	18	fewer	few	ADJ
cana-4260	30	19	samples	sample	NOUN
cana-4260	30	20	than	than	ADP
cana-4260	30	21	the	the	DET
cana-4260	30	22	majority	majority	NOUN
cana-4260	30	23	class	class	NOUN
cana-4260	30	24	.	.	PUNCT
cana-4260	31	1	the	the	DET
cana-4260	31	2	imbalance	imbalance	NOUN
cana-4260	31	3	in	in	ADP
cana-4260	31	4	this	this	DET
cana-4260	31	5	case	case	NOUN
cana-4260	31	6	results	result	VERB
cana-4260	31	7	in	in	ADP
cana-4260	31	8	biased	biased	ADJ
cana-4260	31	9	models	model	NOUN
cana-4260	31	10	,	,	PUNCT
cana-4260	31	11	where	where	SCONJ
cana-4260	31	12	the	the	DET
cana-4260	31	13	majority	majority	NOUN
cana-4260	31	14	class	class	NOUN
cana-4260	31	15	is	be	AUX
cana-4260	31	16	privileged	privileged	ADJ
cana-4260	31	17	,	,	PUNCT
cana-4260	31	18	leading	lead	VERB
cana-4260	31	19	to	to	ADP
cana-4260	31	20	low	low	ADJ
cana-4260	31	21	performance	performance	NOUN
cana-4260	31	22	in	in	ADP
cana-4260	31	23	most	most	ADJ
cana-4260	31	24	cases	case	NOUN
cana-4260	31	25	.	.	PUNCT
cana-4260	32	1	this	this	DET
cana-4260	32	2	problem	problem	NOUN
cana-4260	32	3	is	be	AUX
cana-4260	32	4	more	more	ADV
cana-4260	32	5	complicated	complicated	ADJ
cana-4260	32	6	in	in	ADP
cana-4260	32	7	high	high	ADJ
cana-4260	32	8	-	-	PUNCT
cana-4260	32	9	dimensional	dimensional	ADJ
cana-4260	32	10	and	and	CCONJ
cana-4260	32	11	complex	complex	ADJ
cana-4260	32	12	datasets	dataset	NOUN
cana-4260	32	13	,	,	PUNCT
cana-4260	32	14	where	where	SCONJ
cana-4260	32	15	learning	learn	VERB
cana-4260	32	16	meaningful	meaningful	ADJ
cana-4260	32	17	patterns	pattern	NOUN
cana-4260	32	18	for	for	ADP
cana-4260	32	19	the	the	DET
cana-4260	32	20	minority	minority	NOUN
cana-4260	32	21	class	class	NOUN
cana-4260	32	22	is	be	AUX
cana-4260	32	23	much	much	ADV
cana-4260	32	24	tougher	tough	ADJ
cana-4260	32	25	,	,	PUNCT
cana-4260	32	26	given	give	VERB
cana-4260	32	27	the	the	DET
cana-4260	32	28	overlapping	overlap	VERB
cana-4260	32	29	data	data	NOUN
cana-4260	32	30	points	point	NOUN
cana-4260	32	31	,	,	PUNCT
cana-4260	32	32	and	and	CCONJ
cana-4260	32	33	standard	standard	ADJ
cana-4260	32	34	performance	performance	NOUN
cana-4260	32	35	metrics	metric	NOUN
cana-4260	32	36	such	such	ADJ
cana-4260	32	37	as	as	ADP
cana-4260	32	38	accuracy	accuracy	NOUN
cana-4260	32	39	can	can	AUX
cana-4260	32	40	not	not	PART
cana-4260	32	41	reflect	reflect	VERB
cana-4260	32	42	the	the	DET
cana-4260	32	43	true	true	ADJ
cana-4260	32	44	effectiveness	effectiveness	NOUN
cana-4260	32	45	of	of	ADP
cana-4260	32	46	the	the	DET
cana-4260	32	47	model	model	NOUN
cana-4260	32	48	.	.	PUNCT
cana-4260	33	1	resampling	resample	VERB
cana-4260	33	2	solutions	solution	NOUN
cana-4260	33	3	is	be	AUX
cana-4260	33	4	a	a	DET
cana-4260	33	5	set	set	NOUN
cana-4260	33	6	of	of	ADP
cana-4260	33	7	techniques	technique	NOUN
cana-4260	33	8	to	to	PART
cana-4260	33	9	address	address	VERB
cana-4260	33	10	the	the	DET
cana-4260	33	11	class	class	NOUN
cana-4260	33	12	imbalance	imbalance	NOUN
cana-4260	33	13	problem	problem	NOUN
cana-4260	33	14	that	that	PRON
cana-4260	33	15	occurs	occur	VERB
cana-4260	33	16	in	in	ADP
cana-4260	33	17	many	many	ADJ
cana-4260	33	18	datasets	dataset	NOUN
cana-4260	33	19	.	.	PUNCT
cana-4260	34	1	class	class	NOUN
cana-4260	34	2	imbalance	imbalance	NOUN
cana-4260	34	3	occurs	occur	VERB
cana-4260	34	4	when	when	SCONJ
cana-4260	34	5	one	one	NUM
cana-4260	34	6	class	class	NOUN
cana-4260	34	7	has	have	VERB
cana-4260	34	8	many	many	ADJ
cana-4260	34	9	more	more	ADJ
cana-4260	34	10	instances	instance	NOUN
cana-4260	34	11	than	than	ADP
cana-4260	34	12	the	the	DET
cana-4260	34	13	others	other	NOUN
cana-4260	34	14	,	,	PUNCT
cana-4260	34	15	making	make	VERB
cana-4260	34	16	the	the	DET
cana-4260	34	17	model	model	NOUN
cana-4260	34	18	biased	bias	VERB
cana-4260	34	19	toward	toward	ADP
cana-4260	34	20	predicting	predict	VERB
cana-4260	34	21	the	the	DET
cana-4260	34	22	majority	majority	NOUN
cana-4260	34	23	class	class	NOUN
cana-4260	34	24	.	.	PUNCT
cana-4260	35	1	resampling	resample	VERB
cana-4260	35	2	solutions	solution	NOUN
cana-4260	35	3	try	try	VERB
cana-4260	35	4	to	to	PART
cana-4260	35	5	modify	modify	VERB
cana-4260	35	6	the	the	DET
cana-4260	35	7	dataset	dataset	NOUN
cana-4260	35	8	through	through	ADP
cana-4260	35	9	oversampling	oversample	VERB
cana-4260	35	10	to	to	PART
cana-4260	35	11	increase	increase	VERB
cana-4260	35	12	the	the	DET
cana-4260	35	13	instances	instance	NOUN
cana-4260	35	14	of	of	ADP
cana-4260	35	15	the	the	DET
cana-4260	35	16	minority	minority	NOUN
cana-4260	35	17	class	class	NOUN
cana-4260	35	18	or	or	CCONJ
cana-4260	35	19	undersampling	undersample	VERB
cana-4260	35	20	to	to	PART
cana-4260	35	21	decrease	decrease	VERB
cana-4260	35	22	the	the	DET
cana-4260	35	23	number	number	NOUN
cana-4260	35	24	of	of	ADP
cana-4260	35	25	instances	instance	NOUN
cana-4260	35	26	of	of	ADP
cana-4260	35	27	the	the	DET
cana-4260	35	28	majority	majority	NOUN
cana-4260	35	29	class	class	NOUN
cana-4260	35	30	to	to	PART
cana-4260	35	31	achieve	achieve	VERB
cana-4260	35	32	better	well	ADJ
cana-4260	35	33	class	class	NOUN
cana-4260	35	34	distribution	distribution	NOUN
cana-4260	35	35	.	.	PUNCT
cana-4260	36	1	such	such	ADJ
cana-4260	36	2	modifications	modification	NOUN
cana-4260	36	3	are	be	AUX
cana-4260	36	4	likely	likely	ADJ
cana-4260	36	5	to	to	PART
cana-4260	36	6	benefit	benefit	VERB
cana-4260	36	7	the	the	DET
cana-4260	36	8	model	model	NOUN
cana-4260	36	9	since	since	SCONJ
cana-4260	36	10	it	it	PRON
cana-4260	36	11	would	would	AUX
cana-4260	36	12	be	be	AUX
cana-4260	36	13	trained	train	VERB
cana-4260	36	14	without	without	ADP
cana-4260	36	15	bias	bias	NOUN
cana-4260	36	16	,	,	PUNCT
cana-4260	36	17	thereby	thereby	ADV
cana-4260	36	18	enhancing	enhance	VERB
cana-4260	36	19	the	the	DET
cana-4260	36	20	capability	capability	NOUN
cana-4260	36	21	of	of	ADP
cana-4260	36	22	giving	give	VERB
cana-4260	36	23	better	well	ADJ
cana-4260	36	24	predictions	prediction	NOUN
cana-4260	36	25	about	about	ADP
cana-4260	36	26	both	both	DET
cana-4260	36	27	classes	class	NOUN
cana-4260	36	28	.	.	PUNCT
cana-4260	37	1	examples	example	NOUN
cana-4260	37	2	of	of	ADP
cana-4260	37	3	resampling	resample	VERB
cana-4260	37	4	solutions	solution	NOUN
cana-4260	37	5	include	include	VERB
cana-4260	37	6	smote	smote	ADJ
cana-4260	37	7	,	,	PUNCT
cana-4260	37	8	random	random	ADJ
cana-4260	37	9	oversampling	oversampling	NOUN
cana-4260	37	10	,	,	PUNCT
cana-4260	37	11	random	random	ADJ
cana-4260	37	12	undersampling	undersampling	ADJ
cana-4260	37	13	,	,	PUNCT
cana-4260	37	14	and	and	CCONJ
cana-4260	37	15	cluster	cluster	NOUN
cana-4260	37	16	-	-	PUNCT
cana-4260	37	17	based	base	VERB
cana-4260	37	18	undersampling	undersampling	ADJ
cana-4260	37	19	.	.	PUNCT
cana-4260	38	1	these	these	DET
cana-4260	38	2	methods	method	NOUN
cana-4260	38	3	can	can	AUX
cana-4260	38	4	result	result	VERB
cana-4260	38	5	in	in	ADP
cana-4260	38	6	good	good	ADJ
cana-4260	38	7	model	model	NOUN
cana-4260	38	8	performance	performance	NOUN
cana-4260	38	9	but	but	CCONJ
cana-4260	38	10	have	have	VERB
cana-4260	38	11	to	to	PART
cana-4260	38	12	be	be	AUX
cana-4260	38	13	used	use	VERB
cana-4260	38	14	with	with	ADP
cana-4260	38	15	caution	caution	NOUN
cana-4260	38	16	because	because	SCONJ
cana-4260	38	17	of	of	ADP
cana-4260	38	18	problems	problem	NOUN
cana-4260	38	19	such	such	ADJ
cana-4260	38	20	as	as	ADP
cana-4260	38	21	overfitting	overfitting	NOUN
cana-4260	38	22	or	or	CCONJ
cana-4260	38	23	loss	loss	NOUN
cana-4260	38	24	of	of	ADP
cana-4260	38	25	important	important	ADJ
cana-4260	38	26	data	datum	NOUN
cana-4260	38	27	.	.	PUNCT
cana-4260	39	1	smote	smote	PROPN
cana-4260	39	2	works	work	VERB
cana-4260	39	3	better	well	ADV
cana-4260	39	4	than	than	ADP
cana-4260	39	5	random	random	ADJ
cana-4260	39	6	undersampling	undersampling	ADJ
cana-4260	39	7	(	(	PUNCT
cana-4260	39	8	rus	rus	NOUN
cana-4260	39	9	)	)	PUNCT
cana-4260	39	10	and	and	CCONJ
cana-4260	39	11	random	random	ADJ
cana-4260	39	12	oversampling	oversampling	NOUN
cana-4260	39	13	(	(	PUNCT
cana-4260	39	14	ros	ros	PROPN
cana-4260	39	15	)	)	PUNCT
cana-4260	39	16	because	because	SCONJ
cana-4260	39	17	it	it	PRON
cana-4260	39	18	does	do	AUX
cana-4260	39	19	not	not	PART
cana-4260	39	20	merely	merely	ADV
cana-4260	39	21	replicate	replicate	VERB
cana-4260	39	22	the	the	DET
cana-4260	39	23	minority	minority	NOUN
cana-4260	39	24	class	class	NOUN
cana-4260	39	25	data	datum	NOUN
cana-4260	39	26	but	but	CCONJ
cana-4260	39	27	creates	create	VERB
cana-4260	39	28	new	new	ADJ
cana-4260	39	29	synthetic	synthetic	ADJ
cana-4260	39	30	instances	instance	NOUN
cana-4260	39	31	by	by	ADP
cana-4260	39	32	interpolating	interpolate	VERB
cana-4260	39	33	between	between	ADP
cana-4260	39	34	existing	exist	VERB
cana-4260	39	35	samples	sample	NOUN
cana-4260	39	36	.	.	PUNCT
cana-4260	40	1	this	this	PRON
cana-4260	40	2	avoids	avoid	VERB
cana-4260	40	3	overfitting	overfitting	NOUN
cana-4260	40	4	,	,	PUNCT
cana-4260	40	5	which	which	PRON
cana-4260	40	6	may	may	AUX
cana-4260	40	7	occur	occur	VERB
cana-4260	40	8	in	in	ADP
cana-4260	40	9	ros	ros	PROPN
cana-4260	40	10	due	due	ADP
cana-4260	40	11	to	to	ADP
cana-4260	40	12	repeated	repeat	VERB
cana-4260	40	13	instances	instance	NOUN
cana-4260	40	14	,	,	PUNCT
cana-4260	40	15	and	and	CCONJ
cana-4260	40	16	retains	retain	VERB
cana-4260	40	17	more	more	ADJ
cana-4260	40	18	information	information	NOUN
cana-4260	40	19	than	than	ADP
cana-4260	40	20	rus	rus	NOUN
cana-4260	40	21	,	,	PUNCT
cana-4260	40	22	which	which	PRON
cana-4260	40	23	discards	discard	VERB
cana-4260	40	24	data	datum	NOUN
cana-4260	40	25	from	from	ADP
cana-4260	40	26	the	the	DET
cana-4260	40	27	majority	majority	NOUN
cana-4260	40	28	class	class	NOUN
cana-4260	40	29	.	.	PUNCT
cana-4260	41	1	communications	communication	NOUN
cana-4260	41	2	on	on	ADP
cana-4260	41	3	applied	apply	VERB
cana-4260	41	4	nonlinear	nonlinear	ADJ
cana-4260	41	5	analysis	analysis	NOUN
cana-4260	41	6	issn	issn	NOUN
cana-4260	41	7	:	:	PUNCT
cana-4260	41	8	1074	1074	NUM
cana-4260	41	9	-	-	PUNCT
cana-4260	41	10	133x	133x	NUM
cana-4260	41	11	vol	vol	NOUN
cana-4260	41	12	32	32	NUM
cana-4260	41	13	no	no	NOUN
cana-4260	41	14	.	.	PUNCT
cana-4260	42	1	9s	9s	NUM
cana-4260	42	2	(	(	PUNCT
cana-4260	42	3	2025	2025	NUM
cana-4260	42	4	)	)	PUNCT
cana-4260	42	5	1632	1632	NUM
cana-4260	43	1	https://internationalpubls.com	https://internationalpubls.com	X
cana-4260	43	2	smote	smote	NOUN
cana-4260	43	3	also	also	ADV
cana-4260	43	4	has	have	VERB
cana-4260	43	5	various	various	ADJ
cana-4260	43	6	variants	variant	NOUN
cana-4260	43	7	,	,	PUNCT
cana-4260	43	8	among	among	ADP
cana-4260	43	9	which	which	PRON
cana-4260	43	10	are	be	AUX
cana-4260	43	11	:	:	PUNCT
cana-4260	43	12	basic	basic	ADJ
cana-4260	43	13	smote	smote	NOUN
cana-4260	43	14	,	,	PUNCT
cana-4260	43	15	it	it	PRON
cana-4260	43	16	selects	select	VERB
cana-4260	43	17	one	one	NUM
cana-4260	43	18	or	or	CCONJ
cana-4260	43	19	more	more	ADJ
cana-4260	43	20	of	of	ADP
cana-4260	43	21	its	its	PRON
cana-4260	43	22	nearest	near	ADJ
cana-4260	43	23	neighbors	neighbor	NOUN
cana-4260	43	24	and	and	CCONJ
cana-4260	43	25	generates	generate	VERB
cana-4260	43	26	synthetic	synthetic	ADJ
cana-4260	43	27	samples	sample	NOUN
cana-4260	43	28	.	.	PUNCT
cana-4260	44	1	this	this	PRON
cana-4260	44	2	gives	give	VERB
cana-4260	44	3	new	new	ADJ
cana-4260	44	4	data	datum	NOUN
cana-4260	44	5	points	point	NOUN
cana-4260	44	6	as	as	ADP
cana-4260	44	7	similar	similar	ADJ
cana-4260	44	8	to	to	ADP
cana-4260	44	9	the	the	DET
cana-4260	44	10	original	original	ADJ
cana-4260	44	11	samples	sample	NOUN
cana-4260	44	12	.	.	PUNCT
cana-4260	45	1	borderline	borderline	NOUN
cana-4260	45	2	-	-	PUNCT
cana-4260	45	3	smote	smote	ADJ
cana-4260	45	4	focuses	focus	VERB
cana-4260	45	5	on	on	ADP
cana-4260	45	6	synthesizing	synthesize	VERB
cana-4260	45	7	the	the	DET
cana-4260	45	8	minority	minority	NOUN
cana-4260	45	9	class	class	NOUN
cana-4260	45	10	examples	example	NOUN
cana-4260	45	11	close	close	ADJ
cana-4260	45	12	to	to	ADP
cana-4260	45	13	the	the	DET
cana-4260	45	14	decision	decision	NOUN
cana-4260	45	15	boundary	boundary	NOUN
cana-4260	45	16	of	of	ADP
cana-4260	45	17	majority	majority	NOUN
cana-4260	45	18	versus	versus	ADP
cana-4260	45	19	minority	minority	NOUN
cana-4260	45	20	classes	class	NOUN
cana-4260	45	21	.	.	PUNCT
cana-4260	46	1	it	it	PRON
cana-4260	46	2	produces	produce	VERB
cana-4260	46	3	more	more	ADV
cana-4260	46	4	informative	informative	ADJ
cana-4260	46	5	samples	sample	NOUN
cana-4260	46	6	,	,	PUNCT
cana-4260	46	7	and	and	CCONJ
cana-4260	46	8	that	that	PRON
cana-4260	46	9	improves	improve	VERB
cana-4260	46	10	the	the	DET
cana-4260	46	11	discrimination	discrimination	NOUN
cana-4260	46	12	capability	capability	NOUN
cana-4260	46	13	of	of	ADP
cana-4260	46	14	the	the	DET
cana-4260	46	15	model	model	NOUN
cana-4260	46	16	.	.	PUNCT
cana-4260	47	1	adasyn	adasyn	PROPN
cana-4260	47	2	is	be	AUX
cana-4260	47	3	an	an	DET
cana-4260	47	4	extension	extension	NOUN
cana-4260	47	5	of	of	ADP
cana-4260	47	6	smote	smote	NOUN
cana-4260	47	7	that	that	PRON
cana-4260	47	8	adapts	adapt	VERB
cana-4260	47	9	to	to	ADP
cana-4260	47	10	the	the	DET
cana-4260	47	11	difficulty	difficulty	NOUN
cana-4260	47	12	level	level	NOUN
cana-4260	47	13	of	of	ADP
cana-4260	47	14	classifying	classify	VERB
cana-4260	47	15	samples.it	samples.it	NUM
cana-4260	47	16	focuses	focus	VERB
cana-4260	47	17	on	on	ADP
cana-4260	47	18	generating	generate	VERB
cana-4260	47	19	challenging	challenging	ADJ
cana-4260	47	20	examples	example	NOUN
cana-4260	47	21	,	,	PUNCT
cana-4260	47	22	which	which	PRON
cana-4260	47	23	can	can	AUX
cana-4260	47	24	improve	improve	VERB
cana-4260	47	25	the	the	DET
cana-4260	47	26	model	model	NOUN
cana-4260	47	27	’s	’s	PART
cana-4260	47	28	ability	ability	NOUN
cana-4260	47	29	to	to	PART
cana-4260	47	30	handle	handle	VERB
cana-4260	47	31	complex	complex	ADJ
cana-4260	47	32	decision	decision	NOUN
cana-4260	47	33	boundaries	boundary	NOUN
cana-4260	47	34	in	in	ADP
cana-4260	47	35	this	this	DET
cana-4260	47	36	case	case	NOUN
cana-4260	47	37	,	,	PUNCT
cana-4260	47	38	there	there	PRON
cana-4260	47	39	is	be	VERB
cana-4260	47	40	emphasis	emphasis	NOUN
cana-4260	47	41	on	on	ADP
cana-4260	47	42	generating	generate	VERB
cana-4260	47	43	difficult	difficult	ADJ
cana-4260	47	44	samples	sample	NOUN
cana-4260	47	45	.	.	PUNCT
cana-4260	48	1	that	that	PRON
cana-4260	48	2	may	may	AUX
cana-4260	48	3	enhance	enhance	VERB
cana-4260	48	4	its	its	PRON
cana-4260	48	5	ability	ability	NOUN
cana-4260	48	6	to	to	PART
cana-4260	48	7	classify	classify	VERB
cana-4260	48	8	well	well	ADV
cana-4260	48	9	in	in	ADP
cana-4260	48	10	complex	complex	ADJ
cana-4260	48	11	decision	decision	NOUN
cana-4260	48	12	boundaries	boundary	NOUN
cana-4260	48	13	.	.	PUNCT
cana-4260	49	1	smoteenn	smoteenn	NOUN
cana-4260	49	2	is	be	AUX
cana-4260	49	3	basically	basically	ADV
cana-4260	49	4	the	the	DET
cana-4260	49	5	combination	combination	NOUN
cana-4260	49	6	of	of	ADP
cana-4260	49	7	both	both	DET
cana-4260	49	8	techniques	technique	NOUN
cana-4260	49	9	:	:	PUNCT
cana-4260	49	10	that	that	PRON
cana-4260	49	11	of	of	ADP
cana-4260	49	12	smote	smote	NOUN
cana-4260	49	13	with	with	ADP
cana-4260	49	14	enn	enn	PROPN
cana-4260	49	15	cleaning	cleaning	NOUN
cana-4260	49	16	technique	technique	NOUN
cana-4260	49	17	.	.	PUNCT
cana-4260	50	1	the	the	DET
cana-4260	50	2	method	method	NOUN
cana-4260	50	3	employs	employ	VERB
cana-4260	50	4	enn	enn	PROPN
cana-4260	50	5	right	right	ADV
cana-4260	50	6	after	after	SCONJ
cana-4260	50	7	it	it	PRON
cana-4260	50	8	produced	produce	VERB
cana-4260	50	9	synthetic	synthetic	ADJ
cana-4260	50	10	samples	sample	NOUN
cana-4260	50	11	,	,	PUNCT
cana-4260	50	12	to	to	PART
cana-4260	50	13	eliminate	eliminate	VERB
cana-4260	50	14	the	the	DET
cana-4260	50	15	noisy	noisy	ADJ
cana-4260	50	16	instances	instance	NOUN
cana-4260	50	17	possibly	possibly	ADV
cana-4260	50	18	coming	come	VERB
cana-4260	50	19	from	from	ADP
cana-4260	50	20	a	a	DET
cana-4260	50	21	minority	minority	NOUN
cana-4260	50	22	or	or	CCONJ
cana-4260	50	23	even	even	ADV
cana-4260	50	24	majority	majority	NOUN
cana-4260	50	25	class	class	NOUN
cana-4260	50	26	.	.	PUNCT
cana-4260	51	1	this	this	DET
cana-4260	51	2	results	result	VERB
cana-4260	51	3	in	in	ADP
cana-4260	51	4	an	an	DET
cana-4260	51	5	enhancement	enhancement	NOUN
cana-4260	51	6	of	of	ADP
cana-4260	51	7	the	the	DET
cana-4260	51	8	dataset	dataset	ADJ
cana-4260	51	9	quality	quality	NOUN
cana-4260	51	10	as	as	SCONJ
cana-4260	51	11	misclassified	misclassified	ADJ
cana-4260	51	12	or	or	CCONJ
cana-4260	51	13	borderline	borderline	NOUN
cana-4260	51	14	cases	case	NOUN
cana-4260	51	15	are	be	AUX
cana-4260	51	16	eliminated	eliminate	VERB
cana-4260	51	17	.	.	PUNCT
cana-4260	52	1	in	in	ADP
cana-4260	52	2	smotetomek	smotetomek	PROPN
cana-4260	52	3	,	,	PUNCT
cana-4260	52	4	the	the	DET
cana-4260	52	5	techniques	technique	NOUN
cana-4260	52	6	involve	involve	VERB
cana-4260	52	7	combining	combine	VERB
cana-4260	52	8	smote	smote	NOUN
cana-4260	52	9	with	with	ADP
cana-4260	52	10	tomek	tomek	PROPN
cana-4260	52	11	links	link	NOUN
cana-4260	52	12	cleaning	clean	VERB
cana-4260	52	13	method	method	NOUN
cana-4260	52	14	.	.	PUNCT
cana-4260	53	1	apart	apart	ADV
cana-4260	53	2	from	from	ADP
cana-4260	53	3	balancing	balance	VERB
cana-4260	53	4	the	the	DET
cana-4260	53	5	dataset	dataset	NOUN
cana-4260	53	6	,	,	PUNCT
cana-4260	53	7	it	it	PRON
cana-4260	53	8	also	also	ADV
cana-4260	53	9	removes	remove	VERB
cana-4260	53	10	ambiguities	ambiguity	NOUN
cana-4260	53	11	because	because	SCONJ
cana-4260	53	12	it	it	PRON
cana-4260	53	13	removes	remove	VERB
cana-4260	53	14	those	those	DET
cana-4260	53	15	pairs	pair	NOUN
cana-4260	53	16	that	that	PRON
cana-4260	53	17	will	will	AUX
cana-4260	53	18	probably	probably	ADV
cana-4260	53	19	be	be	AUX
cana-4260	53	20	misclassified	misclassifie	VERB
cana-4260	53	21	.	.	PUNCT
cana-4260	54	1	each	each	DET
cana-4260	54	2	variant	variant	NOUN
cana-4260	54	3	increases	increase	VERB
cana-4260	54	4	the	the	DET
cana-4260	54	5	performance	performance	NOUN
cana-4260	54	6	by	by	ADP
cana-4260	54	7	tackling	tackle	VERB
cana-4260	54	8	particular	particular	ADJ
cana-4260	54	9	problems	problem	NOUN
cana-4260	54	10	,	,	PUNCT
cana-4260	54	11	and	and	CCONJ
cana-4260	54	12	borderline	borderline	NOUN
cana-4260	54	13	-	-	PUNCT
cana-4260	54	14	smote	smote	ADJ
cana-4260	54	15	would	would	AUX
cana-4260	54	16	perform	perform	VERB
cana-4260	54	17	better	well	ADJ
cana-4260	54	18	than	than	ADP
cana-4260	54	19	basic	basic	ADJ
cana-4260	54	20	smote	smote	NOUN
cana-4260	54	21	when	when	SCONJ
cana-4260	54	22	there	there	PRON
cana-4260	54	23	are	be	VERB
cana-4260	54	24	a	a	DET
cana-4260	54	25	lot	lot	NOUN
cana-4260	54	26	of	of	ADP
cana-4260	54	27	borderline	borderline	NOUN
cana-4260	54	28	instances	instance	NOUN
cana-4260	54	29	in	in	ADP
cana-4260	54	30	the	the	DET
cana-4260	54	31	dataset	dataset	NOUN
cana-4260	54	32	,	,	PUNCT
cana-4260	54	33	whereas	whereas	SCONJ
cana-4260	54	34	adasyn	adasyn	PROPN
cana-4260	54	35	could	could	AUX
cana-4260	54	36	work	work	VERB
cana-4260	54	37	better	well	ADV
cana-4260	54	38	with	with	ADP
cana-4260	54	39	large	large	ADJ
cana-4260	54	40	datasets	dataset	NOUN
cana-4260	54	41	where	where	SCONJ
cana-4260	54	42	most	most	ADJ
cana-4260	54	43	samples	sample	NOUN
cana-4260	54	44	are	be	AUX
cana-4260	54	45	hard	hard	ADJ
cana-4260	54	46	to	to	PART
cana-4260	54	47	classify	classify	VERB
cana-4260	54	48	.	.	PUNCT
cana-4260	55	1	smoteenn	smoteenn	NOUN
cana-4260	55	2	and	and	CCONJ
cana-4260	55	3	smotetomek	smotetomek	PROPN
cana-4260	55	4	are	be	AUX
cana-4260	55	5	more	more	ADV
cana-4260	55	6	applicable	applicable	ADJ
cana-4260	55	7	in	in	ADP
cana-4260	55	8	noisy	noisy	ADJ
cana-4260	55	9	or	or	CCONJ
cana-4260	55	10	ambiguous	ambiguous	ADJ
cana-4260	55	11	cases	case	NOUN
cana-4260	55	12	because	because	SCONJ
cana-4260	55	13	they	they	PRON
cana-4260	55	14	improve	improve	VERB
cana-4260	55	15	the	the	DET
cana-4260	55	16	generalization	generalization	NOUN
cana-4260	55	17	capability	capability	NOUN
cana-4260	55	18	of	of	ADP
cana-4260	55	19	the	the	DET
cana-4260	55	20	model	model	NOUN
cana-4260	55	21	.	.	PUNCT
cana-4260	56	1	in	in	ADP
cana-4260	56	2	conclusion	conclusion	NOUN
cana-4260	56	3	,	,	PUNCT
cana-4260	56	4	smote	smote	ADJ
cana-4260	56	5	and	and	CCONJ
cana-4260	56	6	its	its	PRON
cana-4260	56	7	variants	variant	NOUN
cana-4260	56	8	are	be	AUX
cana-4260	56	9	mostly	mostly	ADV
cana-4260	56	10	used	use	VERB
cana-4260	56	11	to	to	ADP
cana-4260	56	12	improvement	improvement	NOUN
cana-4260	56	13	of	of	ADP
cana-4260	56	14	the	the	DET
cana-4260	56	15	performance	performance	NOUN
cana-4260	56	16	of	of	ADP
cana-4260	56	17	classifiers	classifier	NOUN
cana-4260	56	18	on	on	ADP
cana-4260	56	19	imbalanced	imbalanced	ADJ
cana-4260	56	20	data	datum	NOUN
cana-4260	56	21	,	,	PUNCT
cana-4260	56	22	or	or	CCONJ
cana-4260	56	23	better	well	ADJ
cana-4260	56	24	management	management	NOUN
cana-4260	56	25	of	of	ADP
cana-4260	56	26	the	the	DET
cana-4260	56	27	minority	minority	NOUN
cana-4260	56	28	class	class	NOUN
cana-4260	56	29	and	and	CCONJ
cana-4260	56	30	more	more	ADV
cana-4260	56	31	accurate	accurate	ADJ
cana-4260	56	32	overall	overall	ADJ
cana-4260	56	33	models	model	NOUN
cana-4260	56	34	.	.	PUNCT
cana-4260	57	1	2	2	X
cana-4260	57	2	.	.	X
cana-4260	57	3	literature	literature	NOUN
cana-4260	57	4	survey	survey	PROPN
cana-4260	57	5	nitesh	nitesh	ADV
cana-4260	57	6	et	et	PROPN
cana-4260	57	7	al	al	PROPN
cana-4260	57	8	.	.	PROPN
cana-4260	57	9	presented	present	VERB
cana-4260	57	10	smote	smote	PROPN
cana-4260	57	11	,	,	PUNCT
cana-4260	57	12	which	which	PRON
cana-4260	57	13	is	be	AUX
cana-4260	57	14	an	an	DET
cana-4260	57	15	algorithm	algorithm	NOUN
cana-4260	57	16	that	that	PRON
cana-4260	57	17	tries	try	VERB
cana-4260	57	18	to	to	PART
cana-4260	57	19	solve	solve	VERB
cana-4260	57	20	class	class	NOUN
cana-4260	57	21	imbalance	imbalance	NOUN
cana-4260	57	22	in	in	ADP
cana-4260	57	23	machine	machine	NOUN
cana-4260	57	24	learning	learn	VERB
cana-4260	57	25	by	by	ADP
cana-4260	57	26	synthesizing	synthesize	VERB
cana-4260	57	27	new	new	ADJ
cana-4260	57	28	samples	sample	NOUN
cana-4260	57	29	for	for	ADP
cana-4260	57	30	the	the	DET
cana-4260	57	31	minority	minority	NOUN
cana-4260	57	32	class	class	NOUN
cana-4260	57	33	.	.	PUNCT
cana-4260	58	1	it	it	PRON
cana-4260	58	2	does	do	AUX
cana-4260	58	3	not	not	PART
cana-4260	58	4	repeat	repeat	VERB
cana-4260	58	5	minority	minority	NOUN
cana-4260	58	6	samples	sample	NOUN
cana-4260	58	7	,	,	PUNCT
cana-4260	58	8	like	like	ADP
cana-4260	58	9	traditional	traditional	ADJ
cana-4260	58	10	random	random	ADJ
cana-4260	58	11	over	over	ADP
cana-4260	58	12	-	-	PUNCT
cana-4260	58	13	sampling	sampling	NOUN
cana-4260	58	14	does	doe	NOUN
cana-4260	58	15	,	,	PUNCT
cana-4260	58	16	thus	thus	ADV
cana-4260	58	17	it	it	PRON
cana-4260	58	18	is	be	AUX
cana-4260	58	19	less	less	ADV
cana-4260	58	20	likely	likely	ADJ
cana-4260	58	21	to	to	PART
cana-4260	58	22	overfit	overfit	VERB
cana-4260	58	23	.	.	PUNCT
cana-4260	59	1	this	this	PRON
cana-4260	59	2	makes	make	VERB
cana-4260	59	3	it	it	PRON
cana-4260	59	4	improve	improve	VERB
cana-4260	59	5	variability	variability	NOUN
cana-4260	59	6	within	within	ADP
cana-4260	59	7	the	the	DET
cana-4260	59	8	dataset	dataset	NOUN
cana-4260	59	9	,	,	PUNCT
cana-4260	59	10	enhance	enhance	VERB
cana-4260	59	11	generalization	generalization	NOUN
cana-4260	59	12	of	of	ADP
cana-4260	59	13	the	the	DET
cana-4260	59	14	classifier	classifier	NOUN
cana-4260	59	15	,	,	PUNCT
cana-4260	59	16	and	and	CCONJ
cana-4260	59	17	substantially	substantially	ADV
cana-4260	59	18	improve	improve	VERB
cana-4260	59	19	metrics	metric	NOUN
cana-4260	59	20	like	like	ADP
cana-4260	59	21	recall	recall	NOUN
cana-4260	59	22	,	,	PUNCT
cana-4260	59	23	precision	precision	NOUN
cana-4260	59	24	,	,	PUNCT
cana-4260	59	25	and	and	CCONJ
cana-4260	59	26	f1	f1	NOUN
cana-4260	59	27	-	-	PUNCT
cana-4260	59	28	score	score	NOUN
cana-4260	59	29	for	for	ADP
cana-4260	59	30	the	the	DET
cana-4260	59	31	minority	minority	NOUN
cana-4260	59	32	class	class	NOUN
cana-4260	59	33	.	.	PUNCT
cana-4260	60	1	the	the	DET
cana-4260	60	2	paper	paper	NOUN
cana-4260	60	3	does	do	AUX
cana-4260	60	4	report	report	VERB
cana-4260	60	5	significant	significant	ADJ
cana-4260	60	6	improvement	improvement	NOUN
cana-4260	60	7	in	in	ADP
cana-4260	60	8	accuracy	accuracy	NOUN
cana-4260	60	9	performance	performance	NOUN
cana-4260	60	10	on	on	ADP
cana-4260	60	11	several	several	ADJ
cana-4260	60	12	data	datum	NOUN
cana-4260	60	13	sets	set	NOUN
cana-4260	60	14	compared	compare	VERB
cana-4260	60	15	with	with	ADP
cana-4260	60	16	baseline	baseline	ADJ
cana-4260	60	17	methods	method	NOUN
cana-4260	60	18	,	,	PUNCT
cana-4260	60	19	specially	specially	ADV
cana-4260	60	20	for	for	ADP
cana-4260	60	21	decision	decision	NOUN
cana-4260	60	22	tree	tree	NOUN
cana-4260	60	23	classification	classification	NOUN
cana-4260	60	24	scenarios	scenario	NOUN
cana-4260	60	25	.	.	PUNCT
cana-4260	61	1	smote	smote	PROPN
cana-4260	61	2	focused	focus	VERB
cana-4260	61	3	mainly	mainly	ADV
cana-4260	61	4	on	on	ADP
cana-4260	61	5	handling	handle	VERB
cana-4260	61	6	imbalanced	imbalanced	ADJ
cana-4260	61	7	data	data	NOUN
cana-4260	61	8	sets	set	NOUN
cana-4260	61	9	in	in	ADP
cana-4260	61	10	order	order	NOUN
cana-4260	61	11	to	to	PART
cana-4260	61	12	find	find	VERB
cana-4260	61	13	minority	minority	NOUN
cana-4260	61	14	classes	class	NOUN
cana-4260	61	15	better	well	ADV
cana-4260	61	16	and	and	CCONJ
cana-4260	61	17	applied	apply	VERB
cana-4260	61	18	in	in	ADP
cana-4260	61	19	fraud	fraud	NOUN
cana-4260	61	20	detection	detection	NOUN
cana-4260	61	21	,	,	PUNCT
cana-4260	61	22	medical	medical	ADJ
cana-4260	61	23	diagnostics	diagnostic	NOUN
cana-4260	61	24	,	,	PUNCT
cana-4260	61	25	and	and	CCONJ
cana-4260	61	26	fault	fault	VERB
cana-4260	61	27	detection	detection	NOUN
cana-4260	61	28	domains	domain	NOUN
cana-4260	61	29	,	,	PUNCT
cana-4260	61	30	among	among	ADP
cana-4260	61	31	others	other	NOUN
cana-4260	61	32	.	.	PUNCT
cana-4260	62	1	the	the	DET
cana-4260	62	2	benefits	benefit	NOUN
cana-4260	62	3	are	be	AUX
cana-4260	62	4	that	that	SCONJ
cana-4260	62	5	the	the	DET
cana-4260	62	6	models	model	NOUN
cana-4260	62	7	improve	improve	VERB
cana-4260	62	8	in	in	ADP
cana-4260	62	9	performance	performance	NOUN
cana-4260	62	10	,	,	PUNCT
cana-4260	62	11	decrease	decrease	VERB
cana-4260	62	12	the	the	DET
cana-4260	62	13	chances	chance	NOUN
cana-4260	62	14	of	of	ADP
cana-4260	62	15	overfitting	overfitte	VERB
cana-4260	62	16	,	,	PUNCT
cana-4260	62	17	and	and	CCONJ
cana-4260	62	18	are	be	AUX
cana-4260	62	19	flexible	flexible	ADJ
cana-4260	62	20	enough	enough	ADV
cana-4260	62	21	to	to	PART
cana-4260	62	22	support	support	VERB
cana-4260	62	23	any	any	DET
cana-4260	62	24	machine	machine	NOUN
cana-4260	62	25	learning	learn	VERB
cana-4260	62	26	algorithm	algorithm	NOUN
cana-4260	62	27	.	.	PUNCT
cana-4260	63	1	its	its	PRON
cana-4260	63	2	innovative	innovative	ADJ
cana-4260	63	3	aspect	aspect	NOUN
cana-4260	63	4	is	be	AUX
cana-4260	63	5	generating	generate	VERB
cana-4260	63	6	diverse	diverse	ADJ
cana-4260	63	7	synthetic	synthetic	ADJ
cana-4260	63	8	data	datum	NOUN
cana-4260	63	9	,	,	PUNCT
cana-4260	63	10	hence	hence	ADV
cana-4260	63	11	why	why	SCONJ
cana-4260	63	12	the	the	DET
cana-4260	63	13	technique	technique	NOUN
cana-4260	63	14	has	have	AUX
cana-4260	63	15	gained	gain	VERB
cana-4260	63	16	popularity	popularity	NOUN
cana-4260	63	17	for	for	ADP
cana-4260	63	18	being	be	AUX
cana-4260	63	19	used	use	VERB
cana-4260	63	20	on	on	ADP
cana-4260	63	21	imbalanced	imbalanced	ADJ
cana-4260	63	22	datasets[1	datasets[1	ADV
cana-4260	63	23	]	]	PUNCT
cana-4260	63	24	.	.	PUNCT
cana-4260	64	1	haibo	haibo	VERB
cana-4260	64	2	he	he	PRON
cana-4260	64	3	et	et	PROPN
cana-4260	64	4	al	al	PROPN
cana-4260	64	5	.	.	PROPN
cana-4260	64	6	proposed	propose	VERB
cana-4260	64	7	an	an	DET
cana-4260	64	8	extension	extension	NOUN
cana-4260	64	9	of	of	ADP
cana-4260	64	10	smote	smote	NOUN
cana-4260	64	11	called	call	VERB
cana-4260	64	12	adasyn	adasyn	PROPN
cana-4260	64	13	,	,	PUNCT
cana-4260	64	14	which	which	PRON
cana-4260	64	15	will	will	AUX
cana-4260	64	16	adaptively	adaptively	ADV
cana-4260	64	17	generate	generate	VERB
cana-4260	64	18	synthetic	synthetic	ADJ
cana-4260	64	19	samples	sample	NOUN
cana-4260	64	20	for	for	ADP
cana-4260	64	21	the	the	DET
cana-4260	64	22	minority	minority	NOUN
cana-4260	64	23	class	class	NOUN
cana-4260	64	24	based	base	VERB
cana-4260	64	25	on	on	ADP
cana-4260	64	26	data	datum	NOUN
cana-4260	64	27	distribution	distribution	NOUN
cana-4260	64	28	.	.	PUNCT
cana-4260	65	1	unlike	unlike	ADP
cana-4260	65	2	smote	smote	NOUN
cana-4260	65	3	,	,	PUNCT
cana-4260	65	4	adasyn	adasyn	NOUN
cana-4260	65	5	focuses	focus	VERB
cana-4260	65	6	on	on	ADP
cana-4260	65	7	creating	create	VERB
cana-4260	65	8	more	more	ADJ
cana-4260	65	9	synthetic	synthetic	ADJ
cana-4260	65	10	samples	sample	NOUN
cana-4260	65	11	for	for	ADP
cana-4260	65	12	harder	hard	ADJ
cana-4260	65	13	-	-	PUNCT
cana-4260	65	14	to	to	PART
cana-4260	65	15	-	-	PUNCT
cana-4260	65	16	learn	learn	VERB
cana-4260	65	17	instances	instance	NOUN
cana-4260	65	18	,	,	PUNCT
cana-4260	65	19	such	such	ADJ
cana-4260	65	20	as	as	ADP
cana-4260	65	21	those	those	PRON
cana-4260	65	22	near	near	ADP
cana-4260	65	23	the	the	DET
cana-4260	65	24	class	class	NOUN
cana-4260	65	25	boundary	boundary	NOUN
cana-4260	65	26	while	while	SCONJ
cana-4260	65	27	creating	create	VERB
cana-4260	65	28	fewer	few	ADJ
cana-4260	65	29	synthetic	synthetic	ADJ
cana-4260	65	30	samples	sample	NOUN
cana-4260	65	31	for	for	ADP
cana-4260	65	32	well	well	ADV
cana-4260	65	33	-	-	PUNCT
cana-4260	65	34	represented	represent	VERB
cana-4260	65	35	regions	region	NOUN
cana-4260	65	36	.	.	PUNCT
cana-4260	66	1	this	this	PRON
cana-4260	66	2	reduces	reduce	VERB
cana-4260	66	3	bias	bias	NOUN
cana-4260	66	4	in	in	ADP
cana-4260	66	5	imbalanced	imbalanced	ADJ
cana-4260	66	6	datasets	dataset	NOUN
cana-4260	66	7	and	and	CCONJ
cana-4260	66	8	shifts	shift	NOUN
cana-4260	66	9	the	the	DET
cana-4260	66	10	decision	decision	NOUN
cana-4260	66	11	boundary	boundary	NOUN
cana-4260	66	12	of	of	ADP
cana-4260	66	13	the	the	DET
cana-4260	66	14	classifiers	classifier	NOUN
cana-4260	66	15	towards	towards	ADP
cana-4260	66	16	the	the	DET
cana-4260	66	17	minority	minority	NOUN
cana-4260	66	18	communications	communication	NOUN
cana-4260	66	19	on	on	ADP
cana-4260	66	20	applied	apply	VERB
cana-4260	66	21	nonlinear	nonlinear	ADJ
cana-4260	66	22	analysis	analysis	NOUN
cana-4260	66	23	issn	issn	NOUN
cana-4260	66	24	:	:	PUNCT
cana-4260	66	25	1074	1074	NUM
cana-4260	66	26	-	-	PUNCT
cana-4260	66	27	133x	133x	NUM
cana-4260	66	28	vol	vol	NOUN
cana-4260	66	29	32	32	NUM
cana-4260	67	1	no	no	NOUN
cana-4260	67	2	.	.	PUNCT
cana-4260	68	1	9s	9s	NUM
cana-4260	68	2	(	(	PUNCT
cana-4260	68	3	2025	2025	NUM
cana-4260	68	4	)	)	PUNCT
cana-4260	68	5	1633	1633	NUM
cana-4260	68	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-4260	68	7	class	class	NOUN
cana-4260	68	8	dynamically	dynamically	ADV
cana-4260	68	9	.	.	PUNCT
cana-4260	69	1	the	the	DET
cana-4260	69	2	advantage	advantage	NOUN
cana-4260	69	3	of	of	ADP
cana-4260	69	4	adasyn	adasyn	NOUN
cana-4260	69	5	is	be	AUX
cana-4260	69	6	mainly	mainly	ADV
cana-4260	69	7	the	the	DET
cana-4260	69	8	adaptively	adaptively	ADV
cana-4260	69	9	refinement	refinement	NOUN
cana-4260	69	10	of	of	ADP
cana-4260	69	11	the	the	DET
cana-4260	69	12	dataset	dataset	NOUN
cana-4260	69	13	where	where	SCONJ
cana-4260	69	14	focus	focus	NOUN
cana-4260	69	15	is	be	AUX
cana-4260	69	16	made	make	VERB
cana-4260	69	17	upon	upon	SCONJ
cana-4260	69	18	difficult	difficult	ADJ
cana-4260	69	19	regions	region	NOUN
cana-4260	69	20	.	.	PUNCT
cana-4260	70	1	the	the	DET
cana-4260	70	2	overall	overall	ADJ
cana-4260	70	3	performance	performance	NOUN
cana-4260	70	4	on	on	ADP
cana-4260	70	5	imbalanced	imbalanced	ADJ
cana-4260	70	6	datasets	dataset	NOUN
cana-4260	70	7	is	be	AUX
cana-4260	70	8	improved	improve	VERB
cana-4260	70	9	because	because	SCONJ
cana-4260	70	10	adasyn	adasyn	PROPN
cana-4260	70	11	improves	improve	VERB
cana-4260	70	12	the	the	DET
cana-4260	70	13	classifier	classifier	NOUN
cana-4260	70	14	’s	’s	PART
cana-4260	70	15	ability	ability	NOUN
cana-4260	70	16	to	to	PART
cana-4260	70	17	classify	classify	VERB
cana-4260	70	18	the	the	DET
cana-4260	70	19	minority	minority	NOUN
cana-4260	70	20	class	class	NOUN
cana-4260	70	21	instances	instance	NOUN
cana-4260	70	22	and	and	CCONJ
cana-4260	70	23	enhances	enhance	VERB
cana-4260	70	24	generalization[2	generalization[2	PROPN
cana-4260	70	25	]	]	PUNCT
cana-4260	70	26	.	.	PUNCT
cana-4260	71	1	yuchun	yuchun	PROPN
cana-4260	71	2	tang	tang	PROPN
cana-4260	71	3	et	et	PROPN
cana-4260	71	4	al	al	PROPN
cana-4260	71	5	.	.	PROPN
cana-4260	71	6	worked	work	VERB
cana-4260	71	7	in	in	ADP
cana-4260	71	8	the	the	DET
cana-4260	71	9	direction	direction	NOUN
cana-4260	71	10	to	to	PART
cana-4260	71	11	enhance	enhance	VERB
cana-4260	71	12	support	support	NOUN
cana-4260	71	13	vector	vector	NOUN
cana-4260	71	14	machines	machine	NOUN
cana-4260	71	15	(	(	PUNCT
cana-4260	71	16	svms	svms	NOUN
cana-4260	71	17	)	)	PUNCT
cana-4260	71	18	in	in	ADP
cana-4260	71	19	dealing	deal	VERB
cana-4260	71	20	with	with	ADP
cana-4260	71	21	highly	highly	ADV
cana-4260	71	22	imbalanced	imbalanced	ADJ
cana-4260	71	23	datasets	dataset	NOUN
cana-4260	71	24	.	.	PUNCT
cana-4260	72	1	modifications	modification	NOUN
cana-4260	72	2	to	to	ADP
cana-4260	72	3	the	the	DET
cana-4260	72	4	traditional	traditional	ADJ
cana-4260	72	5	svms	svms	NOUN
cana-4260	72	6	,	,	PUNCT
cana-4260	72	7	cost	cost	NOUN
cana-4260	72	8	-	-	PUNCT
cana-4260	72	9	sensitive	sensitive	ADJ
cana-4260	72	10	learning	learning	NOUN
cana-4260	72	11	altering	alter	VERB
cana-4260	72	12	misclassification	misclassification	NOUN
cana-4260	72	13	costs	cost	NOUN
cana-4260	72	14	for	for	ADP
cana-4260	72	15	the	the	DET
cana-4260	72	16	minority	minority	NOUN
cana-4260	72	17	class	class	NOUN
cana-4260	72	18	,	,	PUNCT
cana-4260	72	19	and	and	CCONJ
cana-4260	72	20	the	the	DET
cana-4260	72	21	application	application	NOUN
cana-4260	72	22	of	of	ADP
cana-4260	72	23	oversampling	oversample	VERB
cana-4260	72	24	and	and	CCONJ
cana-4260	72	25	undersampling	undersample	VERB
cana-4260	72	26	to	to	PART
cana-4260	72	27	balance	balance	NOUN
cana-4260	72	28	data	datum	NOUN
cana-4260	72	29	also	also	ADV
cana-4260	72	30	were	be	AUX
cana-4260	72	31	proposed	propose	VERB
cana-4260	72	32	.	.	PUNCT
cana-4260	73	1	an	an	DET
cana-4260	73	2	innovation	innovation	NOUN
cana-4260	73	3	is	be	AUX
cana-4260	73	4	through	through	ADP
cana-4260	73	5	the	the	DET
cana-4260	73	6	gsvm	gsvm	NOUN
cana-4260	73	7	-	-	PUNCT
cana-4260	73	8	ru	ru	NOUN
cana-4260	73	9	,	,	PUNCT
cana-4260	73	10	granularized	granularize	VERB
cana-4260	73	11	version	version	NOUN
cana-4260	73	12	of	of	ADP
cana-4260	73	13	an	an	DET
cana-4260	73	14	svm	svm	PROPN
cana-4260	73	15	,	,	PUNCT
cana-4260	73	16	based	base	VERB
cana-4260	73	17	on	on	ADP
cana-4260	73	18	information	information	NOUN
cana-4260	73	19	loss	loss	NOUN
cana-4260	73	20	occurring	occur	VERB
cana-4260	73	21	when	when	SCONJ
cana-4260	73	22	data	datum	NOUN
cana-4260	73	23	cleaned	clean	VERB
cana-4260	73	24	has	have	VERB
cana-4260	73	25	lesser	less	ADJ
cana-4260	73	26	support	support	NOUN
cana-4260	73	27	vector	vector	NOUN
cana-4260	73	28	count	count	NOUN
cana-4260	73	29	for	for	ADP
cana-4260	73	30	increasing	increase	VERB
cana-4260	73	31	predictions	prediction	NOUN
cana-4260	73	32	speed	speed	VERB
cana-4260	73	33	to	to	ADP
cana-4260	73	34	better	well	ADJ
cana-4260	73	35	model	model	NOUN
cana-4260	73	36	quality	quality	NOUN
cana-4260	73	37	.	.	PUNCT
cana-4260	74	1	they	they	PRON
cana-4260	74	2	improve	improve	VERB
cana-4260	74	3	better	well	ADJ
cana-4260	74	4	performances	performance	NOUN
cana-4260	74	5	over	over	ADP
cana-4260	74	6	svm	svm	ADJ
cana-4260	74	7	approaches	approach	NOUN
cana-4260	74	8	concerning	concern	VERB
cana-4260	74	9	the	the	DET
cana-4260	74	10	capacity	capacity	NOUN
cana-4260	74	11	of	of	ADP
cana-4260	74	12	detection	detection	NOUN
cana-4260	74	13	from	from	ADP
cana-4260	74	14	a	a	DET
cana-4260	74	15	minority	minority	NOUN
cana-4260	74	16	-	-	PUNCT
cana-4260	74	17	class	class	NOUN
cana-4260	74	18	instances	instance	NOUN
cana-4260	74	19	and	and	CCONJ
cana-4260	74	20	exhibits	exhibit	VERB
cana-4260	74	21	high	high	ADJ
cana-4260	74	22	performance	performance	NOUN
cana-4260	74	23	superiority	superiority	NOUN
cana-4260	74	24	compared	compare	VERB
cana-4260	74	25	to	to	ADP
cana-4260	74	26	baselines	baseline	NOUN
cana-4260	74	27	on	on	ADP
cana-4260	74	28	benchmark	benchmark	NOUN
cana-4260	74	29	datasets	dataset	NOUN
cana-4260	74	30	and	and	CCONJ
cana-4260	74	31	metrics	metric	NOUN
cana-4260	74	32	using	use	VERB
cana-4260	74	33	gmean	gmean	ADJ
cana-4260	74	34	and	and	CCONJ
cana-4260	74	35	auc	auc	VERB
cana-4260	74	36	evaluation	evaluation	NOUN
cana-4260	74	37	along	along	ADP
cana-4260	74	38	with	with	ADP
cana-4260	74	39	metrics	metric	NOUN
cana-4260	74	40	precision	precision	NOUN
cana-4260	74	41	/	/	SYM
cana-4260	74	42	recall	recall	NOUN
cana-4260	74	43	measures[3	measures[3	PROPN
cana-4260	74	44	]	]	PUNCT
cana-4260	74	45	.	.	PUNCT
cana-4260	75	1	hui	hui	PROPN
cana-4260	75	2	han	han	PROPN
cana-4260	75	3	et	et	PROPN
cana-4260	75	4	al	al	PROPN
cana-4260	75	5	.	.	PROPN
cana-4260	75	6	proposed	propose	VERB
cana-4260	75	7	borderline	borderline	NOUN
cana-4260	75	8	-	-	PUNCT
cana-4260	75	9	smote	smote	ADJ
cana-4260	75	10	,	,	PUNCT
cana-4260	75	11	an	an	DET
cana-4260	75	12	advanced	advanced	ADJ
cana-4260	75	13	version	version	NOUN
cana-4260	75	14	of	of	ADP
cana-4260	75	15	smote	smote	NOUN
cana-4260	75	16	to	to	PART
cana-4260	75	17	boost	boost	VERB
cana-4260	75	18	the	the	DET
cana-4260	75	19	performance	performance	NOUN
cana-4260	75	20	of	of	ADP
cana-4260	75	21	classifiers	classifier	NOUN
cana-4260	75	22	on	on	ADP
cana-4260	75	23	imbalanced	imbalanced	ADJ
cana-4260	75	24	datasets	dataset	NOUN
cana-4260	75	25	.	.	PUNCT
cana-4260	76	1	the	the	DET
cana-4260	76	2	key	key	ADJ
cana-4260	76	3	aim	aim	NOUN
cana-4260	76	4	is	be	AUX
cana-4260	76	5	to	to	PART
cana-4260	76	6	overcome	overcome	VERB
cana-4260	76	7	the	the	DET
cana-4260	76	8	weakness	weakness	NOUN
cana-4260	76	9	of	of	ADP
cana-4260	76	10	smote	smote	NOUN
cana-4260	76	11	,	,	PUNCT
cana-4260	76	12	which	which	PRON
cana-4260	76	13	is	be	AUX
cana-4260	76	14	the	the	DET
cana-4260	76	15	generation	generation	NOUN
cana-4260	76	16	of	of	ADP
cana-4260	76	17	synthetic	synthetic	ADJ
cana-4260	76	18	samples	sample	NOUN
cana-4260	76	19	only	only	ADV
cana-4260	76	20	for	for	ADP
cana-4260	76	21	the	the	DET
cana-4260	76	22	minority	minority	NOUN
cana-4260	76	23	class	class	NOUN
cana-4260	76	24	instances	instance	NOUN
cana-4260	76	25	near	near	ADP
cana-4260	76	26	the	the	DET
cana-4260	76	27	decision	decision	NOUN
cana-4260	76	28	boundary	boundary	NOUN
cana-4260	76	29	,	,	PUNCT
cana-4260	76	30	or	or	CCONJ
cana-4260	76	31	the	the	DET
cana-4260	76	32	”	"	PUNCT
cana-4260	76	33	borderline	borderline	NOUN
cana-4260	76	34	”	"	PUNCT
cana-4260	76	35	points	point	NOUN
cana-4260	76	36	,	,	PUNCT
cana-4260	76	37	which	which	PRON
cana-4260	76	38	are	be	AUX
cana-4260	76	39	harder	hard	ADJ
cana-4260	76	40	to	to	PART
cana-4260	76	41	classify	classify	VERB
cana-4260	76	42	.	.	PUNCT
cana-4260	77	1	this	this	PRON
cana-4260	77	2	helps	help	VERB
cana-4260	77	3	borderline	borderline	NOUN
cana-4260	77	4	-	-	PUNCT
cana-4260	77	5	smote	smote	ADJ
cana-4260	77	6	improve	improve	VERB
cana-4260	77	7	classifier	classifier	NOUN
cana-4260	77	8	accuracy	accuracy	NOUN
cana-4260	77	9	by	by	ADP
cana-4260	77	10	over	over	ADV
cana-4260	77	11	-	-	PUNCT
cana-4260	77	12	sampling	sample	VERB
cana-4260	77	13	selectively	selectively	ADV
cana-4260	77	14	all	all	DET
cana-4260	77	15	borderline	borderline	NOUN
cana-4260	77	16	instances	instance	NOUN
cana-4260	77	17	;	;	PUNCT
cana-4260	77	18	the	the	DET
cana-4260	77	19	model	model	NOUN
cana-4260	77	20	gets	get	VERB
cana-4260	77	21	a	a	DET
cana-4260	77	22	good	good	ADJ
cana-4260	77	23	capturing	capturing	NOUN
cana-4260	77	24	of	of	ADP
cana-4260	77	25	the	the	DET
cana-4260	77	26	decision	decision	NOUN
cana-4260	77	27	boundary	boundary	ADJ
cana-4260	77	28	between	between	ADP
cana-4260	77	29	classes	class	NOUN
cana-4260	77	30	.	.	PUNCT
cana-4260	78	1	they	they	PRON
cana-4260	78	2	did	do	VERB
cana-4260	78	3	a	a	DET
cana-4260	78	4	major	major	ADJ
cana-4260	78	5	alteration	alteration	NOUN
cana-4260	78	6	by	by	ADP
cana-4260	78	7	border	border	NOUN
cana-4260	78	8	-	-	PUNCT
cana-4260	78	9	instance	instance	NOUN
cana-4260	78	10	identification	identification	NOUN
cana-4260	78	11	within	within	ADP
cana-4260	78	12	the	the	DET
cana-4260	78	13	minority	minority	NOUN
cana-4260	78	14	class	class	NOUN
cana-4260	78	15	for	for	ADP
cana-4260	78	16	generating	generate	VERB
cana-4260	78	17	synthetic	synthetic	ADJ
cana-4260	78	18	data	datum	NOUN
cana-4260	78	19	only	only	ADV
cana-4260	78	20	,	,	PUNCT
cana-4260	78	21	in	in	ADP
cana-4260	78	22	contrast	contrast	NOUN
cana-4260	78	23	to	to	ADP
cana-4260	78	24	the	the	DET
cana-4260	78	25	uniform	uniform	NOUN
cana-4260	78	26	sampling	sampling	NOUN
cana-4260	78	27	performed	perform	VERB
cana-4260	78	28	on	on	ADP
cana-4260	78	29	all	all	DET
cana-4260	78	30	points	point	NOUN
cana-4260	78	31	of	of	ADP
cana-4260	78	32	the	the	DET
cana-4260	78	33	minority	minority	NOUN
cana-4260	78	34	.	.	PUNCT
cana-4260	79	1	borderline	borderline	NOUN
cana-4260	79	2	-	-	PUNCT
cana-4260	79	3	smote	smote	ADJ
cana-4260	79	4	advantages	advantage	NOUN
cana-4260	79	5	include	include	VERB
cana-4260	79	6	improved	improve	VERB
cana-4260	79	7	model	model	NOUN
cana-4260	79	8	performance	performance	NOUN
cana-4260	79	9	,	,	PUNCT
cana-4260	79	10	especially	especially	ADV
cana-4260	79	11	in	in	ADP
cana-4260	79	12	identifying	identify	VERB
cana-4260	79	13	minority	minority	NOUN
cana-4260	79	14	class	class	NOUN
cana-4260	79	15	instances	instance	NOUN
cana-4260	79	16	,	,	PUNCT
cana-4260	79	17	increased	increase	VERB
cana-4260	79	18	recall	recall	NOUN
cana-4260	79	19	and	and	CCONJ
cana-4260	79	20	f	f	NOUN
cana-4260	79	21	-	-	PUNCT
cana-4260	79	22	measure	measure	NOUN
cana-4260	79	23	,	,	PUNCT
cana-4260	79	24	and	and	CCONJ
cana-4260	79	25	decreased	decrease	VERB
cana-4260	79	26	overfitting	overfitte	VERB
cana-4260	79	27	,	,	PUNCT
cana-4260	79	28	since	since	SCONJ
cana-4260	79	29	it	it	PRON
cana-4260	79	30	focuses	focus	VERB
cana-4260	79	31	the	the	DET
cana-4260	79	32	sampling	sample	VERB
cana-4260	79	33	process	process	NOUN
cana-4260	79	34	on	on	ADP
cana-4260	79	35	the	the	DET
cana-4260	79	36	most	most	ADV
cana-4260	79	37	critical	critical	ADJ
cana-4260	79	38	instances	instance	NOUN
cana-4260	79	39	near	near	ADP
cana-4260	79	40	the	the	DET
cana-4260	79	41	decision	decision	NOUN
cana-4260	79	42	boundary[4	boundary[4	PROPN
cana-4260	79	43	]	]	PUNCT
cana-4260	79	44	.	.	PUNCT
cana-4260	80	1	felix	felix	PROPN
cana-4260	80	2	last	last	ADJ
cana-4260	80	3	et	et	PROPN
cana-4260	80	4	al	al	PROPN
cana-4260	80	5	.	.	PROPN
cana-4260	81	1	combined	combine	VERB
cana-4260	81	2	k	k	PROPN
cana-4260	81	3	-	-	PUNCT
cana-4260	81	4	means	means	NOUN
cana-4260	81	5	clustering	cluster	VERB
cana-4260	81	6	with	with	ADP
cana-4260	81	7	the	the	DET
cana-4260	81	8	smote	smote	ADJ
cana-4260	81	9	technique	technique	NOUN
cana-4260	81	10	in	in	ADP
cana-4260	81	11	order	order	NOUN
cana-4260	81	12	to	to	PART
cana-4260	81	13	actually	actually	ADV
cana-4260	81	14	improve	improve	VERB
cana-4260	81	15	oversampling	oversample	VERB
cana-4260	81	16	in	in	ADP
cana-4260	81	17	imbalanced	imbalanced	ADJ
cana-4260	81	18	learning	learning	NOUN
cana-4260	81	19	.	.	PUNCT
cana-4260	82	1	the	the	DET
cana-4260	82	2	main	main	ADJ
cana-4260	82	3	focus	focus	NOUN
cana-4260	82	4	here	here	ADV
cana-4260	82	5	would	would	AUX
cana-4260	82	6	be	be	AUX
cana-4260	82	7	the	the	DET
cana-4260	82	8	proposal	proposal	NOUN
cana-4260	82	9	of	of	ADP
cana-4260	82	10	a	a	DET
cana-4260	82	11	new	new	ADJ
cana-4260	82	12	oversampling	oversampling	ADJ
cana-4260	82	13	method	method	NOUN
cana-4260	82	14	that	that	PRON
cana-4260	82	15	incorporates	incorporate	VERB
cana-4260	82	16	clustering	clustering	ADJ
cana-4260	82	17	techniques	technique	NOUN
cana-4260	82	18	for	for	ADP
cana-4260	82	19	enhancing	enhance	VERB
cana-4260	82	20	the	the	DET
cana-4260	82	21	generation	generation	NOUN
cana-4260	82	22	of	of	ADP
cana-4260	82	23	synthetic	synthetic	ADJ
cana-4260	82	24	samples	sample	NOUN
cana-4260	82	25	for	for	ADP
cana-4260	82	26	the	the	DET
cana-4260	82	27	minority	minority	NOUN
cana-4260	82	28	class	class	NOUN
cana-4260	82	29	.	.	PUNCT
cana-4260	83	1	this	this	PRON
cana-4260	83	2	will	will	AUX
cana-4260	83	3	be	be	AUX
cana-4260	83	4	helpful	helpful	ADJ
cana-4260	83	5	in	in	ADP
cana-4260	83	6	dealing	deal	VERB
cana-4260	83	7	with	with	ADP
cana-4260	83	8	the	the	DET
cana-4260	83	9	problem	problem	NOUN
cana-4260	83	10	of	of	ADP
cana-4260	83	11	overfitting	overfitte	VERB
cana-4260	83	12	as	as	SCONJ
cana-4260	83	13	smote	smote	ADJ
cana-4260	83	14	and	and	CCONJ
cana-4260	83	15	k	k	NOUN
cana-4260	83	16	-	-	PUNCT
cana-4260	83	17	means	means	NOUN
cana-4260	83	18	would	would	AUX
cana-4260	83	19	decrease	decrease	VERB
cana-4260	83	20	the	the	DET
cana-4260	83	21	possibility	possibility	NOUN
cana-4260	83	22	of	of	ADP
cana-4260	83	23	producing	produce	VERB
cana-4260	83	24	redundant	redundant	ADJ
cana-4260	83	25	or	or	CCONJ
cana-4260	83	26	too	too	ADV
cana-4260	83	27	similar	similar	ADJ
cana-4260	83	28	synthetic	synthetic	ADJ
cana-4260	83	29	samples	sample	NOUN
cana-4260	83	30	for	for	ADP
cana-4260	83	31	the	the	DET
cana-4260	83	32	highly	highly	ADV
cana-4260	83	33	imbalanced	imbalanced	ADJ
cana-4260	83	34	dataset	dataset	NOUN
cana-4260	83	35	.	.	PUNCT
cana-4260	84	1	the	the	DET
cana-4260	84	2	key	key	ADJ
cana-4260	84	3	modification	modification	NOUN
cana-4260	84	4	is	be	AUX
cana-4260	84	5	the	the	DET
cana-4260	84	6	integration	integration	NOUN
cana-4260	84	7	of	of	ADP
cana-4260	84	8	k	k	PROPN
cana-4260	84	9	-	-	PUNCT
cana-4260	84	10	means	means	NOUN
cana-4260	84	11	clustering	clustering	NOUN
cana-4260	84	12	,	,	PUNCT
cana-4260	84	13	which	which	PRON
cana-4260	84	14	clusters	cluster	VERB
cana-4260	84	15	the	the	DET
cana-4260	84	16	minority	minority	NOUN
cana-4260	84	17	class	class	NOUN
cana-4260	84	18	instances	instance	NOUN
cana-4260	84	19	into	into	ADP
cana-4260	84	20	clusters	cluster	NOUN
cana-4260	84	21	,	,	PUNCT
cana-4260	84	22	followed	follow	VERB
cana-4260	84	23	by	by	ADP
cana-4260	84	24	applying	apply	VERB
cana-4260	84	25	smote	smote	NOUN
cana-4260	84	26	within	within	ADP
cana-4260	84	27	these	these	DET
cana-4260	84	28	clusters	cluster	NOUN
cana-4260	84	29	to	to	PART
cana-4260	84	30	produce	produce	VERB
cana-4260	84	31	synthetic	synthetic	ADJ
cana-4260	84	32	data	datum	NOUN
cana-4260	84	33	.	.	PUNCT
cana-4260	85	1	this	this	DET
cana-4260	85	2	approach	approach	NOUN
cana-4260	85	3	has	have	VERB
cana-4260	85	4	better	well	ADJ
cana-4260	85	5	diversity	diversity	NOUN
cana-4260	85	6	in	in	ADP
cana-4260	85	7	synthetic	synthetic	ADJ
cana-4260	85	8	samples	sample	NOUN
cana-4260	85	9	,	,	PUNCT
cana-4260	85	10	reduces	reduce	VERB
cana-4260	85	11	overfitting	overfitte	VERB
cana-4260	85	12	,	,	PUNCT
cana-4260	85	13	and	and	CCONJ
cana-4260	85	14	improves	improve	VERB
cana-4260	85	15	model	model	NOUN
cana-4260	85	16	performance	performance	NOUN
cana-4260	85	17	,	,	PUNCT
cana-4260	85	18	since	since	SCONJ
cana-4260	85	19	oversampling	oversample	VERB
cana-4260	85	20	is	be	AUX
cana-4260	85	21	more	more	ADV
cana-4260	85	22	representative	representative	NOUN
cana-4260	85	23	of	of	ADP
cana-4260	85	24	the	the	DET
cana-4260	85	25	underlying	underlie	VERB
cana-4260	85	26	distribution	distribution	NOUN
cana-4260	85	27	of	of	ADP
cana-4260	85	28	the	the	DET
cana-4260	85	29	minority	minority	NOUN
cana-4260	85	30	class	class	NOUN
cana-4260	85	31	.	.	PUNCT
cana-4260	86	1	it	it	PRON
cana-4260	86	2	also	also	ADV
cana-4260	86	3	could	could	AUX
cana-4260	86	4	help	help	VERB
cana-4260	86	5	to	to	PART
cana-4260	86	6	enhance	enhance	VERB
cana-4260	86	7	the	the	DET
cana-4260	86	8	classification	classification	NOUN
cana-4260	86	9	for	for	ADP
cana-4260	86	10	the	the	DET
cana-4260	86	11	minority	minority	NOUN
cana-4260	86	12	class	class	NOUN
cana-4260	86	13	instances	instance	NOUN
cana-4260	86	14	without	without	ADP
cana-4260	86	15	over	over	ADP
cana-4260	86	16	-	-	PUNCT
cana-4260	86	17	redundancy	redundancy	NOUN
cana-4260	86	18	in	in	ADP
cana-4260	86	19	synthetic	synthetic	ADJ
cana-4260	86	20	data[5	data[5	NOUN
cana-4260	86	21	]	]	PUNCT
cana-4260	86	22	.	.	PUNCT
cana-4260	87	1	mimi	mimi	PROPN
cana-4260	87	2	mukherjee	mukherjee	PROPN
cana-4260	87	3	and	and	CCONJ
cana-4260	87	4	matloob	matloob	PROPN
cana-4260	87	5	khushi	khushi	PROPN
cana-4260	87	6	proposed	propose	VERB
cana-4260	87	7	an	an	DET
cana-4260	87	8	extension	extension	NOUN
cana-4260	87	9	of	of	ADP
cana-4260	87	10	the	the	DET
cana-4260	87	11	traditional	traditional	ADJ
cana-4260	87	12	smote	smote	ADJ
cana-4260	87	13	technique	technique	NOUN
cana-4260	87	14	,	,	PUNCT
cana-4260	87	15	namely	namely	ADV
cana-4260	87	16	smote	smote	ADJ
cana-4260	87	17	-	-	PUNCT
cana-4260	87	18	enc	enc	NOUN
cana-4260	87	19	,	,	PUNCT
cana-4260	87	20	to	to	PART
cana-4260	87	21	deal	deal	VERB
cana-4260	87	22	with	with	ADP
cana-4260	87	23	datasets	dataset	NOUN
cana-4260	87	24	having	have	VERB
cana-4260	87	25	a	a	DET
cana-4260	87	26	combination	combination	NOUN
cana-4260	87	27	of	of	ADP
cana-4260	87	28	both	both	CCONJ
cana-4260	87	29	nominal	nominal	ADJ
cana-4260	87	30	and	and	CCONJ
cana-4260	87	31	continuous	continuous	ADJ
cana-4260	87	32	features	feature	NOUN
cana-4260	87	33	.	.	PUNCT
cana-4260	88	1	this	this	DET
cana-4260	88	2	paper	paper	NOUN
cana-4260	88	3	is	be	AUX
cana-4260	88	4	primarily	primarily	ADV
cana-4260	88	5	on	on	ADP
cana-4260	88	6	the	the	DET
cana-4260	88	7	limitations	limitation	NOUN
cana-4260	88	8	of	of	ADP
cana-4260	88	9	smote	smote	NOUN
cana-4260	88	10	as	as	ADP
cana-4260	88	11	the	the	DET
cana-4260	88	12	technique	technique	NOUN
cana-4260	88	13	mainly	mainly	ADV
cana-4260	88	14	works	work	VERB
cana-4260	88	15	for	for	ADP
cana-4260	88	16	continuous	continuous	ADJ
cana-4260	88	17	data	datum	NOUN
cana-4260	88	18	,	,	PUNCT
cana-4260	88	19	generating	generate	VERB
cana-4260	88	20	a	a	DET
cana-4260	88	21	method	method	NOUN
cana-4260	88	22	which	which	PRON
cana-4260	88	23	could	could	AUX
cana-4260	88	24	generate	generate	VERB
cana-4260	88	25	synthetic	synthetic	ADJ
cana-4260	88	26	samples	sample	NOUN
cana-4260	88	27	for	for	ADP
cana-4260	88	28	datasets	dataset	NOUN
cana-4260	88	29	containing	contain	VERB
cana-4260	88	30	both	both	DET
cana-4260	88	31	types	type	NOUN
cana-4260	88	32	of	of	ADP
cana-4260	88	33	attributes	attribute	NOUN
cana-4260	88	34	:	:	PUNCT
cana-4260	88	35	nominal	nominal	ADJ
cana-4260	88	36	and	and	CCONJ
cana-4260	88	37	continuous	continuous	ADJ
cana-4260	88	38	.	.	PUNCT
cana-4260	89	1	the	the	DET
cana-4260	89	2	use	use	NOUN
cana-4260	89	3	of	of	ADP
cana-4260	89	4	smote	smote	ADJ
cana-4260	89	5	-	-	PUNCT
cana-4260	89	6	enc	enc	NOUN
cana-4260	89	7	aims	aim	VERB
cana-4260	89	8	to	to	PART
cana-4260	89	9	enhance	enhance	VERB
cana-4260	89	10	classification	classification	NOUN
cana-4260	89	11	performance	performance	NOUN
cana-4260	89	12	in	in	ADP
cana-4260	89	13	imbalanced	imbalanced	ADJ
cana-4260	89	14	datasets	dataset	NOUN
cana-4260	89	15	with	with	ADP
cana-4260	89	16	categorical	categorical	ADJ
cana-4260	89	17	features	feature	NOUN
cana-4260	89	18	by	by	ADP
cana-4260	89	19	creating	create	VERB
cana-4260	89	20	valid	valid	ADJ
cana-4260	89	21	synthetic	synthetic	ADJ
cana-4260	89	22	instances	instance	NOUN
cana-4260	89	23	,	,	PUNCT
cana-4260	89	24	which	which	PRON
cana-4260	89	25	preserve	preserve	VERB
cana-4260	89	26	the	the	DET
cana-4260	89	27	integrity	integrity	NOUN
cana-4260	89	28	of	of	ADP
cana-4260	89	29	both	both	DET
cana-4260	89	30	types	type	NOUN
cana-4260	89	31	of	of	ADP
cana-4260	89	32	features	feature	NOUN
cana-4260	89	33	.	.	PUNCT
cana-4260	90	1	the	the	DET
cana-4260	90	2	main	main	ADJ
cana-4260	90	3	alteration	alteration	NOUN
cana-4260	90	4	made	make	VERB
cana-4260	90	5	is	be	AUX
cana-4260	90	6	adapting	adapt	VERB
cana-4260	90	7	smote	smote	NOUN
cana-4260	90	8	to	to	ADP
cana-4260	90	9	nominal	nominal	ADJ
cana-4260	90	10	data	datum	NOUN
cana-4260	90	11	by	by	ADP
cana-4260	90	12	providing	provide	VERB
cana-4260	90	13	a	a	DET
cana-4260	90	14	communications	communication	NOUN
cana-4260	90	15	on	on	ADP
cana-4260	90	16	applied	apply	VERB
cana-4260	90	17	nonlinear	nonlinear	ADJ
cana-4260	90	18	analysis	analysis	NOUN
cana-4260	90	19	issn	issn	NOUN
cana-4260	90	20	:	:	PUNCT
cana-4260	90	21	1074	1074	NUM
cana-4260	90	22	-	-	PUNCT
cana-4260	90	23	133x	133x	NUM
cana-4260	90	24	vol	vol	NOUN
cana-4260	90	25	32	32	NUM
cana-4260	90	26	no	no	NOUN
cana-4260	90	27	.	.	PUNCT
cana-4260	91	1	9s	9s	NUM
cana-4260	91	2	(	(	PUNCT
cana-4260	91	3	2025	2025	NUM
cana-4260	91	4	)	)	PUNCT
cana-4260	91	5	1634	1634	NUM
cana-4260	92	1	https://internationalpubls.com	https://internationalpubls.com	X
cana-4260	92	2	method	method	NOUN
cana-4260	92	3	of	of	ADP
cana-4260	92	4	handling	handle	VERB
cana-4260	92	5	categorical	categorical	ADJ
cana-4260	92	6	features	feature	NOUN
cana-4260	92	7	during	during	ADP
cana-4260	92	8	the	the	DET
cana-4260	92	9	creation	creation	NOUN
cana-4260	92	10	of	of	ADP
cana-4260	92	11	synthetic	synthetic	ADJ
cana-4260	92	12	samples	sample	NOUN
cana-4260	92	13	.	.	PUNCT
cana-4260	93	1	the	the	DET
cana-4260	93	2	advantages	advantage	NOUN
cana-4260	93	3	of	of	ADP
cana-4260	93	4	smote	smote	ADJ
cana-4260	93	5	-	-	PUNCT
cana-4260	93	6	enc	enc	NOUN
cana-4260	93	7	are	be	AUX
cana-4260	93	8	that	that	SCONJ
cana-4260	93	9	it	it	PRON
cana-4260	93	10	can	can	AUX
cana-4260	93	11	handle	handle	VERB
cana-4260	93	12	mixed	mixed	ADJ
cana-4260	93	13	datasets	dataset	NOUN
cana-4260	93	14	,	,	PUNCT
cana-4260	93	15	keeps	keep	VERB
cana-4260	93	16	the	the	DET
cana-4260	93	17	distribution	distribution	NOUN
cana-4260	93	18	of	of	ADP
cana-4260	93	19	nominal	nominal	ADJ
cana-4260	93	20	and	and	CCONJ
cana-4260	93	21	continuous	continuous	ADJ
cana-4260	93	22	features	feature	NOUN
cana-4260	93	23	,	,	PUNCT
cana-4260	93	24	and	and	CCONJ
cana-4260	93	25	improves	improve	VERB
cana-4260	93	26	classifier	classifier	ADJ
cana-4260	93	27	performance	performance	NOUN
cana-4260	93	28	in	in	ADP
cana-4260	93	29	terms	term	NOUN
cana-4260	93	30	of	of	ADP
cana-4260	93	31	both	both	DET
cana-4260	93	32	minority	minority	NOUN
cana-4260	93	33	class	class	NOUN
cana-4260	93	34	detection	detection	NOUN
cana-4260	93	35	and	and	CCONJ
cana-4260	93	36	overall	overall	ADJ
cana-4260	93	37	accuracy	accuracy	NOUN
cana-4260	93	38	without	without	ADP
cana-4260	93	39	distorting	distort	VERB
cana-4260	93	40	the	the	DET
cana-4260	93	41	data	datum	NOUN
cana-4260	93	42	structure.[6	structure.[6	NUM
cana-4260	93	43	]	]	PUNCT
cana-4260	93	44	.	.	PUNCT
cana-4260	94	1	amerah	amerah	PROPN
cana-4260	94	2	alabrah	alabrah	PROPN
cana-4260	94	3	also	also	ADV
cana-4260	94	4	improved	improve	VERB
cana-4260	94	5	the	the	DET
cana-4260	94	6	performance	performance	NOUN
cana-4260	94	7	of	of	ADP
cana-4260	94	8	the	the	DET
cana-4260	94	9	ccf	ccf	PROPN
cana-4260	94	10	detector	detector	NOUN
cana-4260	94	11	on	on	ADP
cana-4260	94	12	handling	handle	VERB
cana-4260	94	13	class	class	NOUN
cana-4260	94	14	imbalances	imbalance	NOUN
cana-4260	94	15	and	and	CCONJ
cana-4260	94	16	normalization	normalization	NOUN
cana-4260	94	17	of	of	ADP
cana-4260	94	18	outliers	outlier	NOUN
cana-4260	94	19	via	via	ADP
cana-4260	94	20	the	the	DET
cana-4260	94	21	interquartile	interquartile	ADJ
cana-4260	94	22	range	range	NOUN
cana-4260	94	23	(	(	PUNCT
cana-4260	94	24	iqr	iqr	NOUN
cana-4260	94	25	)	)	PUNCT
cana-4260	94	26	method	method	NOUN
cana-4260	94	27	.	.	PUNCT
cana-4260	95	1	the	the	DET
cana-4260	95	2	emphasis	emphasis	NOUN
cana-4260	95	3	in	in	ADP
cana-4260	95	4	this	this	DET
cana-4260	95	5	improvement	improvement	NOUN
cana-4260	95	6	is	be	AUX
cana-4260	95	7	towards	towards	ADP
cana-4260	95	8	making	make	VERB
cana-4260	95	9	the	the	DET
cana-4260	95	10	detector	detector	NOUN
cana-4260	95	11	more	more	ADV
cana-4260	95	12	efficient	efficient	ADJ
cana-4260	95	13	and	and	CCONJ
cana-4260	95	14	effective	effective	ADJ
cana-4260	95	15	for	for	ADP
cana-4260	95	16	application	application	NOUN
cana-4260	95	17	on	on	ADP
cana-4260	95	18	datasets	dataset	NOUN
cana-4260	95	19	which	which	PRON
cana-4260	95	20	have	have	VERB
cana-4260	95	21	classes	class	NOUN
cana-4260	95	22	of	of	ADP
cana-4260	95	23	imbalance	imbalance	NOUN
cana-4260	95	24	and	and	CCONJ
cana-4260	95	25	existent	existent	ADJ
cana-4260	95	26	outliers	outlier	NOUN
cana-4260	95	27	whose	whose	DET
cana-4260	95	28	presence	presence	NOUN
cana-4260	95	29	might	might	AUX
cana-4260	95	30	significantly	significantly	ADV
cana-4260	95	31	degrade	degrade	VERB
cana-4260	95	32	the	the	DET
cana-4260	95	33	outcome	outcome	NOUN
cana-4260	95	34	.	.	PUNCT
cana-4260	96	1	the	the	DET
cana-4260	96	2	iqr	iqr	PROPN
cana-4260	96	3	method	method	NOUN
cana-4260	96	4	helps	help	VERB
cana-4260	96	5	eliminate	eliminate	VERB
cana-4260	96	6	outliers	outlier	NOUN
cana-4260	96	7	in	in	ADP
cana-4260	96	8	the	the	DET
cana-4260	96	9	data	datum	NOUN
cana-4260	96	10	and	and	CCONJ
cana-4260	96	11	reduces	reduce	VERB
cana-4260	96	12	their	their	PRON
cana-4260	96	13	effect	effect	NOUN
cana-4260	96	14	,	,	PUNCT
cana-4260	96	15	making	make	VERB
cana-4260	96	16	the	the	DET
cana-4260	96	17	data	datum	NOUN
cana-4260	96	18	cleaner	clean	ADJ
cana-4260	96	19	and	and	CCONJ
cana-4260	96	20	more	more	ADV
cana-4260	96	21	balanced	balanced	ADJ
cana-4260	96	22	in	in	ADP
cana-4260	96	23	favor	favor	NOUN
cana-4260	96	24	of	of	ADP
cana-4260	96	25	the	the	DET
cana-4260	96	26	classifier	classifier	NOUN
cana-4260	96	27	.	.	PUNCT
cana-4260	97	1	the	the	DET
cana-4260	97	2	major	major	ADJ
cana-4260	97	3	change	change	NOUN
cana-4260	97	4	in	in	ADP
cana-4260	97	5	this	this	DET
cana-4260	97	6	paper	paper	NOUN
cana-4260	97	7	was	be	AUX
cana-4260	97	8	the	the	DET
cana-4260	97	9	integration	integration	NOUN
cana-4260	97	10	of	of	ADP
cana-4260	97	11	the	the	DET
cana-4260	97	12	iqr	iqr	PROPN
cana-4260	97	13	method	method	NOUN
cana-4260	97	14	to	to	PART
cana-4260	97	15	remove	remove	VERB
cana-4260	97	16	or	or	CCONJ
cana-4260	97	17	normalize	normalize	VERB
cana-4260	97	18	outliers	outlier	NOUN
cana-4260	97	19	before	before	ADP
cana-4260	97	20	the	the	DET
cana-4260	97	21	applications	application	NOUN
cana-4260	97	22	of	of	ADP
cana-4260	97	23	a	a	DET
cana-4260	97	24	ccf	ccf	NOUN
cana-4260	97	25	detector	detector	NOUN
cana-4260	97	26	on	on	ADP
cana-4260	97	27	the	the	DET
cana-4260	97	28	data	datum	NOUN
cana-4260	97	29	,	,	PUNCT
cana-4260	97	30	which	which	PRON
cana-4260	97	31	improved	improve	VERB
cana-4260	97	32	the	the	DET
cana-4260	97	33	quality	quality	NOUN
cana-4260	97	34	of	of	ADP
cana-4260	97	35	the	the	DET
cana-4260	97	36	data	datum	NOUN
cana-4260	97	37	used	use	VERB
cana-4260	97	38	for	for	ADP
cana-4260	97	39	training	training	NOUN
cana-4260	97	40	.	.	PUNCT
cana-4260	98	1	the	the	DET
cana-4260	98	2	advantage	advantage	NOUN
cana-4260	98	3	of	of	ADP
cana-4260	98	4	this	this	DET
cana-4260	98	5	approach	approach	NOUN
cana-4260	98	6	is	be	AUX
cana-4260	98	7	that	that	SCONJ
cana-4260	98	8	it	it	PRON
cana-4260	98	9	helps	help	VERB
cana-4260	98	10	the	the	DET
cana-4260	98	11	classifier	classifier	NOUN
cana-4260	98	12	to	to	PART
cana-4260	98	13	work	work	VERB
cana-4260	98	14	better	well	ADV
cana-4260	98	15	with	with	ADP
cana-4260	98	16	imbalanced	imbalanced	ADJ
cana-4260	98	17	data	datum	NOUN
cana-4260	98	18	by	by	ADP
cana-4260	98	19	reducing	reduce	VERB
cana-4260	98	20	the	the	DET
cana-4260	98	21	impact	impact	NOUN
cana-4260	98	22	of	of	ADP
cana-4260	98	23	extreme	extreme	ADJ
cana-4260	98	24	values	value	NOUN
cana-4260	98	25	,	,	PUNCT
cana-4260	98	26	and	and	CCONJ
cana-4260	98	27	by	by	ADP
cana-4260	98	28	improving	improve	VERB
cana-4260	98	29	overall	overall	ADJ
cana-4260	98	30	accuracy	accuracy	NOUN
cana-4260	98	31	,	,	PUNCT
cana-4260	98	32	leading	lead	VERB
cana-4260	98	33	to	to	ADP
cana-4260	98	34	a	a	DET
cana-4260	98	35	more	more	ADV
cana-4260	98	36	reliable	reliable	ADJ
cana-4260	98	37	model	model	NOUN
cana-4260	98	38	in	in	ADP
cana-4260	98	39	real	real	ADJ
cana-4260	98	40	-	-	PUNCT
cana-4260	98	41	world	world	NOUN
cana-4260	98	42	applications	application	NOUN
cana-4260	98	43	where	where	SCONJ
cana-4260	98	44	data	datum	NOUN
cana-4260	98	45	is	be	AUX
cana-4260	98	46	often	often	ADV
cana-4260	98	47	messy	messy	ADJ
cana-4260	98	48	and	and	CCONJ
cana-4260	98	49	noisy.[7	noisy.[7	NOUN
cana-4260	98	50	]	]	X
cana-4260	98	51	.	.	PUNCT
cana-4260	99	1	3	3	X
cana-4260	99	2	.	.	X
cana-4260	99	3	methodology	methodology	NOUN
cana-4260	99	4	:	:	PUNCT
cana-4260	99	5	the	the	DET
cana-4260	99	6	approach	approach	NOUN
cana-4260	99	7	outlined	outline	VERB
cana-4260	99	8	in	in	ADP
cana-4260	99	9	this	this	DET
cana-4260	99	10	paper	paper	NOUN
cana-4260	99	11	utilizes	utilize	VERB
cana-4260	99	12	the	the	DET
cana-4260	99	13	gaussian	gaussian	ADJ
cana-4260	99	14	mixture	mixture	NOUN
cana-4260	99	15	model	model	NOUN
cana-4260	99	16	(	(	PUNCT
cana-4260	99	17	gmm	gmm	NOUN
cana-4260	99	18	)	)	PUNCT
cana-4260	99	19	algorithm	algorithm	NOUN
cana-4260	99	20	together	together	ADV
cana-4260	99	21	with	with	ADP
cana-4260	99	22	smote	smote	ADJ
cana-4260	99	23	oversampling	oversampling	NOUN
cana-4260	99	24	in	in	ADP
cana-4260	99	25	order	order	NOUN
cana-4260	99	26	to	to	PART
cana-4260	99	27	rebalance	rebalance	VERB
cana-4260	99	28	highly	highly	ADV
cana-4260	99	29	imbalanced	imbalanced	ADJ
cana-4260	99	30	datasets	dataset	NOUN
cana-4260	99	31	.	.	PUNCT
cana-4260	100	1	it	it	PRON
cana-4260	100	2	does	do	VERB
cana-4260	100	3	an	an	DET
cana-4260	100	4	effective	effective	ADJ
cana-4260	100	5	job	job	NOUN
cana-4260	100	6	of	of	ADP
cana-4260	100	7	preventing	prevent	VERB
cana-4260	100	8	noise	noise	NOUN
cana-4260	100	9	generation	generation	NOUN
cana-4260	100	10	by	by	ADP
cana-4260	100	11	oversampling	oversample	VERB
cana-4260	100	12	only	only	ADV
cana-4260	100	13	in	in	ADP
cana-4260	100	14	safe	safe	ADJ
cana-4260	100	15	regions	region	NOUN
cana-4260	100	16	,	,	PUNCT
cana-4260	100	17	where	where	SCONJ
cana-4260	100	18	the	the	DET
cana-4260	100	19	data	data	NOUN
cana-4260	100	20	is	be	AUX
cana-4260	100	21	better	well	ADV
cana-4260	100	22	structured	structure	VERB
cana-4260	100	23	.	.	PUNCT
cana-4260	101	1	special	special	ADJ
cana-4260	101	2	emphasis	emphasis	NOUN
cana-4260	101	3	is	be	AUX
cana-4260	101	4	laid	lay	VERB
cana-4260	101	5	on	on	ADP
cana-4260	101	6	handling	handle	VERB
cana-4260	101	7	both	both	CCONJ
cana-4260	101	8	between	between	ADP
cana-4260	101	9	-	-	PUNCT
cana-4260	101	10	class	class	NOUN
cana-4260	101	11	imbalance	imbalance	NOUN
cana-4260	101	12	and	and	CCONJ
cana-4260	101	13	within	within	ADP
cana-4260	101	14	-	-	PUNCT
cana-4260	101	15	class	class	NOUN
cana-4260	101	16	imbalance	imbalance	NOUN
cana-4260	101	17	,	,	PUNCT
cana-4260	101	18	tackling	tackle	VERB
cana-4260	101	19	the	the	DET
cana-4260	101	20	small	small	ADJ
cana-4260	101	21	disjuncts	disjunct	NOUN
cana-4260	101	22	problem	problem	NOUN
cana-4260	101	23	by	by	ADP
cana-4260	101	24	inflating	inflate	VERB
cana-4260	101	25	sparse	sparse	ADJ
cana-4260	101	26	minority	minority	NOUN
cana-4260	101	27	regions	region	NOUN
cana-4260	101	28	.	.	PUNCT
cana-4260	102	1	this	this	DET
cana-4260	102	2	approach	approach	NOUN
cana-4260	102	3	is	be	AUX
cana-4260	102	4	readily	readily	ADV
cana-4260	102	5	applicable	applicable	ADJ
cana-4260	102	6	because	because	SCONJ
cana-4260	102	7	of	of	ADP
cana-4260	102	8	the	the	DET
cana-4260	102	9	ease	ease	NOUN
cana-4260	102	10	and	and	CCONJ
cana-4260	102	11	common	common	ADJ
cana-4260	102	12	availability	availability	NOUN
cana-4260	102	13	of	of	ADP
cana-4260	102	14	both	both	PRON
cana-4260	102	15	smote	smote	ADJ
cana-4260	102	16	and	and	CCONJ
cana-4260	102	17	gmm	gmm	NOUN
cana-4260	102	18	.	.	PUNCT
cana-4260	103	1	it	it	PRON
cana-4260	103	2	is	be	AUX
cana-4260	103	3	particularly	particularly	ADV
cana-4260	103	4	distinct	distinct	ADJ
cana-4260	103	5	from	from	ADP
cana-4260	103	6	similar	similar	ADJ
cana-4260	103	7	approaches	approach	NOUN
cana-4260	103	8	not	not	PART
cana-4260	103	9	only	only	ADV
cana-4260	103	10	because	because	SCONJ
cana-4260	103	11	of	of	ADP
cana-4260	103	12	its	its	PRON
cana-4260	103	13	low	low	ADJ
cana-4260	103	14	complexity	complexity	NOUN
cana-4260	103	15	but	but	CCONJ
cana-4260	103	16	also	also	ADV
cana-4260	103	17	because	because	SCONJ
cana-4260	103	18	of	of	ADP
cana-4260	103	19	its	its	PRON
cana-4260	103	20	efficient	efficient	ADJ
cana-4260	103	21	method	method	NOUN
cana-4260	103	22	of	of	ADP
cana-4260	103	23	distributing	distribute	VERB
cana-4260	103	24	synthetic	synthetic	ADJ
cana-4260	103	25	samples	sample	NOUN
cana-4260	103	26	according	accord	VERB
cana-4260	103	27	to	to	ADP
cana-4260	103	28	the	the	DET
cana-4260	103	29	density	density	NOUN
cana-4260	103	30	of	of	ADP
cana-4260	103	31	gaussian	gaussian	ADJ
cana-4260	103	32	components	component	NOUN
cana-4260	103	33	.	.	PUNCT
cana-4260	104	1	3.1	3.1	NUM
cana-4260	104	2	algorithm	algorithm	NOUN
cana-4260	104	3	the	the	DET
cana-4260	104	4	gmm	gmm	NOUN
cana-4260	104	5	-	-	PUNCT
cana-4260	104	6	smote	smote	ADJ
cana-4260	104	7	algorithm	algorithm	NOUN
cana-4260	104	8	involves	involve	VERB
cana-4260	104	9	three	three	NUM
cana-4260	104	10	steps	step	NOUN
cana-4260	104	11	:	:	PUNCT
cana-4260	104	12	clustering	cluster	VERB
cana-4260	104	13	,	,	PUNCT
cana-4260	104	14	filtering	filtering	NOUN
cana-4260	104	15	,	,	PUNCT
cana-4260	104	16	and	and	CCONJ
cana-4260	104	17	oversampling	oversample	VERB
cana-4260	104	18	.	.	PUNCT
cana-4260	105	1	clustering	clustering	NOUN
cana-4260	105	2	is	be	AUX
cana-4260	105	3	performed	perform	VERB
cana-4260	105	4	in	in	ADP
cana-4260	105	5	the	the	DET
cana-4260	105	6	first	first	ADJ
cana-4260	105	7	step	step	NOUN
cana-4260	105	8	where	where	SCONJ
cana-4260	105	9	the	the	DET
cana-4260	105	10	input	input	NOUN
cana-4260	105	11	space	space	NOUN
cana-4260	105	12	is	be	AUX
cana-4260	105	13	partitioned	partition	VERB
cana-4260	105	14	into	into	ADP
cana-4260	105	15	k	k	PROPN
cana-4260	105	16	groups	group	NOUN
cana-4260	105	17	with	with	ADP
cana-4260	105	18	the	the	DET
cana-4260	105	19	help	help	NOUN
cana-4260	105	20	of	of	ADP
cana-4260	105	21	the	the	DET
cana-4260	105	22	gaussian	gaussian	ADJ
cana-4260	105	23	mixture	mixture	NOUN
cana-4260	105	24	model	model	NOUN
cana-4260	105	25	(	(	PUNCT
cana-4260	105	26	gmm	gmm	PROPN
cana-4260	105	27	)	)	PUNCT
cana-4260	105	28	.	.	PUNCT
cana-4260	106	1	filtering	filter	VERB
cana-4260	106	2	is	be	AUX
cana-4260	106	3	the	the	DET
cana-4260	106	4	second	second	ADJ
cana-4260	106	5	step	step	NOUN
cana-4260	106	6	that	that	PRON
cana-4260	106	7	chooses	choose	VERB
cana-4260	106	8	clusters	cluster	NOUN
cana-4260	106	9	to	to	PART
cana-4260	106	10	be	be	AUX
cana-4260	106	11	oversampled	oversample	VERB
cana-4260	106	12	,	,	PUNCT
cana-4260	106	13	keeping	keep	VERB
cana-4260	106	14	the	the	DET
cana-4260	106	15	ones	one	NOUN
cana-4260	106	16	with	with	ADP
cana-4260	106	17	a	a	DET
cana-4260	106	18	high	high	ADJ
cana-4260	106	19	percentage	percentage	NOUN
cana-4260	106	20	of	of	ADP
cana-4260	106	21	minority	minority	NOUN
cana-4260	106	22	class	class	NOUN
cana-4260	106	23	samples	sample	NOUN
cana-4260	106	24	.	.	PUNCT
cana-4260	107	1	it	it	PRON
cana-4260	107	2	further	far	ADV
cana-4260	107	3	assigns	assign	VERB
cana-4260	107	4	the	the	DET
cana-4260	107	5	number	number	NOUN
cana-4260	107	6	of	of	ADP
cana-4260	107	7	synthetic	synthetic	ADJ
cana-4260	107	8	samples	sample	NOUN
cana-4260	107	9	to	to	PART
cana-4260	107	10	be	be	AUX
cana-4260	107	11	created	create	VERB
cana-4260	107	12	,	,	PUNCT
cana-4260	107	13	giving	give	VERB
cana-4260	107	14	a	a	DET
cana-4260	107	15	higher	high	ADJ
cana-4260	107	16	number	number	NOUN
cana-4260	107	17	of	of	ADP
cana-4260	107	18	samples	sample	NOUN
cana-4260	107	19	to	to	ADP
cana-4260	107	20	those	those	DET
cana-4260	107	21	clusters	cluster	NOUN
cana-4260	107	22	where	where	SCONJ
cana-4260	107	23	minority	minority	NOUN
cana-4260	107	24	class	class	NOUN
cana-4260	107	25	samples	sample	NOUN
cana-4260	107	26	are	be	AUX
cana-4260	107	27	scattered	scatter	VERB
cana-4260	107	28	.	.	PUNCT
cana-4260	108	1	lastly	lastly	ADV
cana-4260	108	2	,	,	PUNCT
cana-4260	108	3	during	during	ADP
cana-4260	108	4	the	the	DET
cana-4260	108	5	oversampling	oversampling	ADJ
cana-4260	108	6	process	process	NOUN
cana-4260	108	7	,	,	PUNCT
cana-4260	108	8	smote	smote	NOUN
cana-4260	108	9	is	be	AUX
cana-4260	108	10	invoked	invoke	VERB
cana-4260	108	11	within	within	ADP
cana-4260	108	12	each	each	DET
cana-4260	108	13	identified	identify	VERB
cana-4260	108	14	cluster	cluster	NOUN
cana-4260	108	15	to	to	PART
cana-4260	108	16	realize	realize	VERB
cana-4260	108	17	the	the	DET
cana-4260	108	18	desired	desire	VERB
cana-4260	108	19	minority	minority	NOUN
cana-4260	108	20	and	and	CCONJ
cana-4260	108	21	majority	majority	NOUN
cana-4260	108	22	instances	instance	NOUN
cana-4260	108	23	ratio	ratio	NOUN
cana-4260	108	24	.	.	PUNCT
cana-4260	109	1	input	input	NOUN
cana-4260	109	2	:	:	PUNCT
cana-4260	109	3	begin	begin	VERB
cana-4260	109	4	:	:	PUNCT
cana-4260	109	5	x	x	X
cana-4260	109	6	(	(	PUNCT
cana-4260	109	7	matrix	matrix	NOUN
cana-4260	109	8	of	of	ADP
cana-4260	109	9	observations	observation	NOUN
cana-4260	109	10	)	)	PUNCT
cana-4260	109	11	,	,	PUNCT
cana-4260	109	12	y	y	PROPN
cana-4260	109	13	(	(	PUNCT
cana-4260	109	14	target	target	NOUN
cana-4260	109	15	vector	vector	NOUN
cana-4260	109	16	)	)	PUNCT
cana-4260	109	17	,	,	PUNCT
cana-4260	109	18	n	n	CCONJ
cana-4260	109	19	(	(	PUNCT
cana-4260	109	20	number	number	NOUN
cana-4260	109	21	of	of	ADP
cana-4260	109	22	samples	sample	NOUN
cana-4260	109	23	to	to	PART
cana-4260	109	24	be	be	AUX
cana-4260	109	25	generated	generate	VERB
cana-4260	109	26	)	)	PUNCT
cana-4260	109	27	,	,	PUNCT
cana-4260	109	28	communications	communication	NOUN
cana-4260	109	29	on	on	ADP
cana-4260	109	30	applied	apply	VERB
cana-4260	109	31	nonlinear	nonlinear	ADJ
cana-4260	109	32	analysis	analysis	NOUN
cana-4260	109	33	issn	issn	NOUN
cana-4260	109	34	:	:	PUNCT
cana-4260	109	35	1074	1074	NUM
cana-4260	109	36	-	-	PUNCT
cana-4260	109	37	133x	133x	NUM
cana-4260	109	38	vol	vol	NOUN
cana-4260	109	39	32	32	NUM
cana-4260	110	1	no	no	NOUN
cana-4260	110	2	.	.	PUNCT
cana-4260	111	1	9s	9s	NUM
cana-4260	111	2	(	(	PUNCT
cana-4260	111	3	2025	2025	NUM
cana-4260	111	4	)	)	PUNCT
cana-4260	111	5	1635	1635	NUM
cana-4260	111	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-4260	112	1	k	k	X
cana-4260	112	2	(	(	PUNCT
cana-4260	112	3	number	number	NOUN
cana-4260	112	4	of	of	ADP
cana-4260	112	5	gaussian	gaussian	ADJ
cana-4260	112	6	components	component	NOUN
cana-4260	112	7	found	find	VERB
cana-4260	112	8	by	by	ADP
cana-4260	112	9	gmm	gmm	PROPN
cana-4260	112	10	)	)	PUNCT
cana-4260	112	11	,	,	PUNCT
cana-4260	112	12	irt	irt	PROPN
cana-4260	112	13	(	(	PUNCT
cana-4260	112	14	imbalance	imbalance	NOUN
cana-4260	112	15	ratio	ratio	NOUN
cana-4260	112	16	threshold	threshold	NOUN
cana-4260	112	17	)	)	PUNCT
cana-4260	112	18	,	,	PUNCT
cana-4260	112	19	knn	knn	PROPN
cana-4260	112	20	(	(	PUNCT
cana-4260	112	21	number	number	NOUN
cana-4260	112	22	of	of	ADP
cana-4260	112	23	nearest	near	ADJ
cana-4260	112	24	neighbors	neighbor	NOUN
cana-4260	112	25	considered	consider	VERB
cana-4260	112	26	by	by	ADP
cana-4260	112	27	smote	smote	NOUN
cana-4260	112	28	)	)	PUNCT
cana-4260	112	29	,	,	PUNCT
cana-4260	112	30	de	de	X
cana-4260	112	31	(	(	PUNCT
cana-4260	112	32	exponent	exponent	NOUN
cana-4260	112	33	for	for	ADP
cana-4260	112	34	density	density	NOUN
cana-4260	112	35	computation	computation	NOUN
cana-4260	112	36	,	,	PUNCT
cana-4260	112	37	defaults	default	NOUN
cana-4260	112	38	to	to	ADP
cana-4260	112	39	the	the	DET
cana-4260	112	40	number	number	NOUN
cana-4260	112	41	of	of	ADP
cana-4260	112	42	features	feature	NOUN
cana-4260	112	43	in	in	ADP
cana-4260	112	44	x	x	NOUN
cana-4260	112	45	)	)	PUNCT
cana-4260	112	46	//	//	NUM
cana-4260	112	47	step	step	NOUN
cana-4260	112	48	1	1	NUM
cana-4260	112	49	:	:	PUNCT
cana-4260	112	50	fit	fit	ADJ
cana-4260	112	51	gmm	gmm	NOUN
cana-4260	112	52	to	to	PART
cana-4260	112	53	input	input	VERB
cana-4260	112	54	space	space	NOUN
cana-4260	112	55	and	and	CCONJ
cana-4260	112	56	filter	filter	NOUN
cana-4260	112	57	components	component	NOUN
cana-4260	112	58	with	with	ADP
cana-4260	112	59	more	more	ADJ
cana-4260	112	60	minority	minority	NOUN
cana-4260	112	61	instances	instance	NOUN
cana-4260	112	62	than	than	ADP
cana-4260	112	63	majority	majority	NOUN
cana-4260	112	64	instances	instance	NOUN
cana-4260	112	65	.	.	PUNCT
cana-4260	113	1	components	component	NOUN
cana-4260	113	2	←	←	PROPN
cana-4260	113	3	gmm(x	gmm(x	PROPN
cana-4260	113	4	,	,	PUNCT
cana-4260	113	5	k	k	NOUN
cana-4260	113	6	)	)	PUNCT
cana-4260	113	7	filteredcomponents	filteredcomponent	NOUN
cana-4260	113	8	←	←	PROPN
cana-4260	113	9	{	{	PUNCT
cana-4260	113	10	}	}	PUNCT
cana-4260	113	11	for	for	ADP
cana-4260	113	12	c	c	PROPN
cana-4260	113	13	in	in	ADP
cana-4260	113	14	components	component	NOUN
cana-4260	113	15	do	do	VERB
cana-4260	113	16	imbalanceratio	imbalanceratio	PROPN
cana-4260	113	17	←	←	PROPN
cana-4260	113	18	(	(	PUNCT
cana-4260	113	19	majoritycount(c	majoritycount(c	PROPN
cana-4260	113	20	)	)	PUNCT
cana-4260	113	21	+	+	CCONJ
cana-4260	113	22	1	1	X
cana-4260	113	23	)	)	PUNCT
cana-4260	113	24	/	/	PUNCT
cana-4260	113	25	(	(	PUNCT
cana-4260	113	26	minoritycount(c	minoritycount(c	VERB
cana-4260	113	27	)	)	PUNCT
cana-4260	113	28	+	+	NUM
cana-4260	113	29	1	1	X
cana-4260	113	30	)	)	PUNCT
cana-4260	113	31	if	if	SCONJ
cana-4260	113	32	imbalanceratio	imbalanceratio	PROPN
cana-4260	113	33	<	<	X
cana-4260	113	34	irt	irt	PROPN
cana-4260	113	35	then	then	ADV
cana-4260	113	36	filteredcomponents	filteredcomponent	VERB
cana-4260	113	37	←	←	PROPN
cana-4260	113	38	filteredcomponents	filteredcomponent	NOUN
cana-4260	113	39	u	u	PROPN
cana-4260	113	40	{	{	PUNCT
cana-4260	113	41	c	c	NOUN
cana-4260	113	42	}	}	PUNCT
cana-4260	113	43	end	end	NOUN
cana-4260	113	44	end	end	NOUN
cana-4260	113	45	//	//	PUNCT
cana-4260	113	46	step	step	NOUN
cana-4260	113	47	2	2	NUM
cana-4260	113	48	:	:	PUNCT
cana-4260	113	49	for	for	ADP
cana-4260	113	50	each	each	DET
cana-4260	113	51	filtered	filter	VERB
cana-4260	113	52	component	component	NOUN
cana-4260	113	53	,	,	PUNCT
cana-4260	113	54	compute	compute	VERB
cana-4260	113	55	the	the	DET
cana-4260	113	56	sampling	sample	VERB
cana-4260	113	57	weight	weight	NOUN
cana-4260	113	58	based	base	VERB
cana-4260	113	59	on	on	ADP
cana-4260	113	60	its	its	PRON
cana-4260	113	61	minority	minority	NOUN
cana-4260	113	62	density	density	NOUN
cana-4260	113	63	.	.	PUNCT
cana-4260	114	1	for	for	SCONJ
cana-4260	114	2	f	f	PROPN
cana-4260	114	3	in	in	ADP
cana-4260	114	4	filteredcomponents	filteredcomponent	NOUN
cana-4260	114	5	do	do	VERB
cana-4260	114	6	averageminoritydistance(f	averageminoritydistance(f	NOUN
cana-4260	114	7	)	)	PUNCT
cana-4260	115	1	←	←	PROPN
cana-4260	115	2	mean	mean	NOUN
cana-4260	115	3	(	(	PUNCT
cana-4260	115	4	euclideandistances(f	euclideandistances(f	NOUN
cana-4260	115	5	)	)	PUNCT
cana-4260	115	6	)	)	PUNCT
cana-4260	115	7	densityfactor(f	densityfactor(f	PROPN
cana-4260	115	8	)	)	PUNCT
cana-4260	115	9	←	←	PROPN
cana-4260	115	10	minoritycount(f	minoritycount(f	PROPN
cana-4260	115	11	)	)	PUNCT
cana-4260	115	12	/	/	SYM
cana-4260	115	13	(	(	PUNCT
cana-4260	115	14	averageminoritydistance(f)^de	averageminoritydistance(f)^de	NOUN
cana-4260	116	1	+	+	CCONJ
cana-4260	116	2	1	1	X
cana-4260	116	3	)	)	PUNCT
cana-4260	116	4	sparsityfactor(f	sparsityfactor(f	NOUN
cana-4260	116	5	)	)	PUNCT
cana-4260	116	6	←	←	PROPN
cana-4260	116	7	1	1	NUM
cana-4260	116	8	/	/	SYM
cana-4260	116	9	densityfactor(f	densityfactor(f	PROPN
cana-4260	116	10	)	)	PUNCT
cana-4260	116	11	end	end	VERB
cana-4260	116	12	sparsitysum	sparsitysum	ADJ
cana-4260	116	13	←	←	PROPN
cana-4260	116	14	sum(sparsityfactor(f	sum(sparsityfactor(f	PROPN
cana-4260	116	15	)	)	PUNCT
cana-4260	116	16	for	for	ADP
cana-4260	116	17	f	f	PROPN
cana-4260	116	18	in	in	ADP
cana-4260	116	19	filteredcomponents	filteredcomponent	NOUN
cana-4260	116	20	)	)	PUNCT
cana-4260	116	21	for	for	ADP
cana-4260	116	22	f	f	PROPN
cana-4260	116	23	in	in	ADP
cana-4260	116	24	filteredcomponents	filteredcomponent	NOUN
cana-4260	116	25	do	do	AUX
cana-4260	116	26	samplingweight(f	samplingweight(f	PROPN
cana-4260	116	27	)	)	PUNCT
cana-4260	116	28	←	←	PROPN
cana-4260	116	29	sparsityfactor(f	sparsityfactor(f	PROPN
cana-4260	116	30	)	)	PUNCT
cana-4260	116	31	/	/	SYM
cana-4260	116	32	sparsitysum	sparsitysum	PROPN
cana-4260	116	33	end	end	NOUN
cana-4260	116	34	//	//	PUNCT
cana-4260	116	35	step	step	NOUN
cana-4260	116	36	3	3	NUM
cana-4260	116	37	:	:	PUNCT
cana-4260	116	38	oversample	oversample	NOUN
cana-4260	116	39	each	each	DET
cana-4260	116	40	filtered	filter	VERB
cana-4260	116	41	component	component	NOUN
cana-4260	116	42	using	use	VERB
cana-4260	116	43	smote	smote	NOUN
cana-4260	116	44	.	.	PUNCT
cana-4260	117	1	the	the	DET
cana-4260	117	2	number	number	NOUN
cana-4260	117	3	of	of	ADP
cana-4260	117	4	samples	sample	NOUN
cana-4260	117	5	to	to	PART
cana-4260	117	6	be	be	AUX
cana-4260	117	7	generated	generate	VERB
cana-4260	117	8	is	be	AUX
cana-4260	117	9	computed	compute	VERB
cana-4260	117	10	using	use	VERB
cana-4260	117	11	the	the	DET
cana-4260	117	12	sampling	sample	VERB
cana-4260	117	13	weight	weight	NOUN
cana-4260	117	14	.	.	PUNCT
cana-4260	118	1	generatedsamples	generatedsample	NOUN
cana-4260	118	2	←	←	PROPN
cana-4260	118	3	{	{	PUNCT
cana-4260	118	4	}	}	PUNCT
cana-4260	118	5	for	for	ADP
cana-4260	118	6	f	f	PROPN
cana-4260	118	7	in	in	ADP
cana-4260	118	8	filteredcomponents	filteredcomponent	NOUN
cana-4260	118	9	do	do	AUX
cana-4260	118	10	numberofsamples	numberofsample	NOUN
cana-4260	118	11	←	←	PROPN
cana-4260	118	12	n	n	PROPN
cana-4260	118	13	*	*	PUNCT
cana-4260	118	14	samplingweight(f	samplingweight(f	PROPN
cana-4260	118	15	)	)	PUNCT
cana-4260	118	16	generatedsamples	generatedsample	NOUN
cana-4260	118	17	←	←	PROPN
cana-4260	118	18	generatedsamples	generatedsample	NOUN
cana-4260	118	19	u	u	PROPN
cana-4260	118	20	smote(f	smote(f	PROPN
cana-4260	118	21	,	,	PUNCT
cana-4260	118	22	numberofsamples	numberofsample	NOUN
cana-4260	118	23	,	,	PUNCT
cana-4260	118	24	knn	knn	PROPN
cana-4260	118	25	)	)	PUNCT
cana-4260	118	26	end	end	VERB
cana-4260	118	27	return	return	NOUN
cana-4260	118	28	generatedsamples	generatedsample	NOUN
cana-4260	118	29	endthe	endthe	DET
cana-4260	118	30	gmm	gmm	PROPN
cana-4260	118	31	is	be	AUX
cana-4260	118	32	a	a	DET
cana-4260	118	33	probabilistic	probabilistic	ADJ
cana-4260	118	34	method	method	NOUN
cana-4260	118	35	of	of	ADP
cana-4260	118	36	clustering	clustering	ADJ
cana-4260	118	37	data	datum	NOUN
cana-4260	118	38	,	,	PUNCT
cana-4260	118	39	under	under	ADP
cana-4260	118	40	the	the	DET
cana-4260	118	41	condition	condition	NOUN
cana-4260	118	42	that	that	SCONJ
cana-4260	118	43	the	the	DET
cana-4260	118	44	data	datum	NOUN
cana-4260	118	45	comes	come	VERB
cana-4260	118	46	from	from	ADP
cana-4260	118	47	a	a	DET
cana-4260	118	48	mixture	mixture	NOUN
cana-4260	118	49	of	of	ADP
cana-4260	118	50	multiple	multiple	ADJ
cana-4260	118	51	gaussian	gaussian	ADJ
cana-4260	118	52	distributions	distribution	NOUN
cana-4260	118	53	.	.	PUNCT
cana-4260	119	1	in	in	ADP
cana-4260	119	2	contrast	contrast	NOUN
cana-4260	119	3	to	to	ADP
cana-4260	119	4	kmeans	kmean	NOUN
cana-4260	119	5	,	,	PUNCT
cana-4260	119	6	which	which	PRON
cana-4260	119	7	puts	put	VERB
cana-4260	119	8	every	every	DET
cana-4260	119	9	data	data	NOUN
cana-4260	119	10	point	point	NOUN
cana-4260	119	11	into	into	ADP
cana-4260	119	12	a	a	DET
cana-4260	119	13	single	single	ADJ
cana-4260	119	14	cluster	cluster	NOUN
cana-4260	119	15	,	,	PUNCT
cana-4260	119	16	gmm	gmm	PROPN
cana-4260	119	17	puts	put	VERB
cana-4260	119	18	probabilities	probability	NOUN
cana-4260	119	19	on	on	ADP
cana-4260	119	20	the	the	DET
cana-4260	119	21	data	data	NOUN
cana-4260	119	22	points	point	NOUN
cana-4260	119	23	reflecting	reflect	VERB
cana-4260	119	24	how	how	SCONJ
cana-4260	119	25	likely	likely	ADJ
cana-4260	119	26	they	they	PRON
cana-4260	119	27	are	be	AUX
cana-4260	119	28	to	to	PART
cana-4260	119	29	belong	belong	VERB
cana-4260	119	30	to	to	ADP
cana-4260	119	31	every	every	DET
cana-4260	119	32	gaussian	gaussian	ADJ
cana-4260	119	33	component	component	NOUN
cana-4260	119	34	.	.	PUNCT
cana-4260	120	1	expectation	expectation	NOUN
cana-4260	120	2	-	-	PUNCT
cana-4260	120	3	maximization	maximization	NOUN
cana-4260	120	4	(	(	PUNCT
cana-4260	120	5	em	em	PRON
cana-4260	120	6	)	)	PUNCT
cana-4260	120	7	algorithm	algorithm	NOUN
cana-4260	120	8	is	be	AUX
cana-4260	120	9	employed	employ	VERB
cana-4260	120	10	to	to	PART
cana-4260	120	11	iteratively	iteratively	ADV
cana-4260	120	12	maximize	maximize	VERB
cana-4260	120	13	the	the	DET
cana-4260	120	14	likelihood	likelihood	NOUN
cana-4260	120	15	of	of	ADP
cana-4260	120	16	the	the	DET
cana-4260	120	17	observed	observe	VERB
cana-4260	120	18	data	datum	NOUN
cana-4260	120	19	by	by	ADP
cana-4260	120	20	estimating	estimate	VERB
cana-4260	120	21	the	the	DET
cana-4260	120	22	parameters	parameter	NOUN
cana-4260	120	23	of	of	ADP
cana-4260	120	24	the	the	DET
cana-4260	120	25	gaussian	gaussian	ADJ
cana-4260	120	26	distributions	distribution	NOUN
cana-4260	120	27	.	.	PUNCT
cana-4260	121	1	this	this	DET
cana-4260	121	2	clustering	clustering	ADJ
cana-4260	121	3	technique	technique	NOUN
cana-4260	121	4	is	be	AUX
cana-4260	121	5	capable	capable	ADJ
cana-4260	121	6	of	of	ADP
cana-4260	121	7	flexible	flexible	ADJ
cana-4260	121	8	,	,	PUNCT
cana-4260	121	9	elliptical	elliptical	ADJ
cana-4260	121	10	cluster	cluster	NOUN
cana-4260	121	11	shapes	shape	NOUN
cana-4260	121	12	and	and	CCONJ
cana-4260	121	13	soft	soft	ADJ
cana-4260	121	14	memberships	membership	NOUN
cana-4260	121	15	,	,	PUNCT
cana-4260	121	16	which	which	PRON
cana-4260	121	17	are	be	AUX
cana-4260	121	18	beneficial	beneficial	ADJ
cana-4260	121	19	for	for	ADP
cana-4260	121	20	datasets	dataset	NOUN
cana-4260	121	21	where	where	SCONJ
cana-4260	121	22	clusters	cluster	NOUN
cana-4260	121	23	vary	vary	VERB
cana-4260	121	24	in	in	ADP
cana-4260	121	25	density	density	NOUN
cana-4260	121	26	and	and	CCONJ
cana-4260	121	27	shape	shape	NOUN
cana-4260	121	28	.	.	PUNCT
cana-4260	122	1	communications	communication	NOUN
cana-4260	122	2	on	on	ADP
cana-4260	122	3	applied	apply	VERB
cana-4260	122	4	nonlinear	nonlinear	ADJ
cana-4260	122	5	analysis	analysis	NOUN
cana-4260	122	6	issn	issn	NOUN
cana-4260	122	7	:	:	PUNCT
cana-4260	122	8	1074	1074	NUM
cana-4260	122	9	-	-	PUNCT
cana-4260	122	10	133x	133x	NUM
cana-4260	122	11	vol	vol	NOUN
cana-4260	122	12	32	32	NUM
cana-4260	122	13	no	no	NOUN
cana-4260	122	14	.	.	PUNCT
cana-4260	123	1	9s	9s	NUM
cana-4260	123	2	(	(	PUNCT
cana-4260	123	3	2025	2025	NUM
cana-4260	123	4	)	)	PUNCT
cana-4260	123	5	1636	1636	NUM
cana-4260	123	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-4260	124	1	all	all	DET
cana-4260	124	2	the	the	DET
cana-4260	124	3	hyperparameters	hyperparameter	NOUN
cana-4260	124	4	of	of	ADP
cana-4260	124	5	gmm	gmm	NOUN
cana-4260	124	6	,	,	PUNCT
cana-4260	124	7	including	include	VERB
cana-4260	124	8	the	the	DET
cana-4260	124	9	number	number	NOUN
cana-4260	124	10	of	of	ADP
cana-4260	124	11	components	component	NOUN
cana-4260	124	12	(	(	PUNCT
cana-4260	124	13	k	k	NOUN
cana-4260	124	14	)	)	PUNCT
cana-4260	124	15	and	and	CCONJ
cana-4260	124	16	covariance	covariance	NOUN
cana-4260	124	17	type	type	NOUN
cana-4260	124	18	,	,	PUNCT
cana-4260	124	19	have	have	VERB
cana-4260	124	20	an	an	DET
cana-4260	124	21	effect	effect	NOUN
cana-4260	124	22	on	on	ADP
cana-4260	124	23	the	the	DET
cana-4260	124	24	results	result	NOUN
cana-4260	124	25	of	of	ADP
cana-4260	124	26	clustering	clustering	NOUN
cana-4260	124	27	.	.	PUNCT
cana-4260	125	1	the	the	DET
cana-4260	125	2	choice	choice	NOUN
cana-4260	125	3	of	of	ADP
cana-4260	125	4	a	a	DET
cana-4260	125	5	proper	proper	ADJ
cana-4260	125	6	value	value	NOUN
cana-4260	125	7	for	for	ADP
cana-4260	125	8	k	k	PROPN
cana-4260	125	9	is	be	AUX
cana-4260	125	10	very	very	ADV
cana-4260	125	11	important	important	ADJ
cana-4260	125	12	,	,	PUNCT
cana-4260	125	13	as	as	SCONJ
cana-4260	125	14	it	it	PRON
cana-4260	125	15	impacts	impact	VERB
cana-4260	125	16	the	the	DET
cana-4260	125	17	number	number	NOUN
cana-4260	125	18	of	of	ADP
cana-4260	125	19	minority	minority	NOUN
cana-4260	125	20	clusters	cluster	NOUN
cana-4260	125	21	that	that	PRON
cana-4260	125	22	can	can	AUX
cana-4260	125	23	be	be	AUX
cana-4260	125	24	identified	identify	VERB
cana-4260	125	25	and	and	CCONJ
cana-4260	125	26	oversampled	oversample	VERB
cana-4260	125	27	in	in	ADP
cana-4260	125	28	the	the	DET
cana-4260	125	29	filtering	filter	VERB
cana-4260	125	30	step	step	NOUN
cana-4260	125	31	.	.	PUNCT
cana-4260	126	1	filtering	filter	VERB
cana-4260	126	2	step	step	NOUN
cana-4260	126	3	:	:	PUNCT
cana-4260	126	4	choosing	choose	VERB
cana-4260	126	5	clusters	cluster	NOUN
cana-4260	126	6	to	to	ADP
cana-4260	126	7	oversample	oversample	NOUN
cana-4260	126	8	following	follow	VERB
cana-4260	126	9	clustering	cluster	VERB
cana-4260	126	10	,	,	PUNCT
cana-4260	126	11	the	the	DET
cana-4260	126	12	filtering	filter	VERB
cana-4260	126	13	process	process	NOUN
cana-4260	126	14	determines	determine	NOUN
cana-4260	126	15	which	which	PRON
cana-4260	126	16	clusters	cluster	NOUN
cana-4260	126	17	have	have	VERB
cana-4260	126	18	a	a	DET
cana-4260	126	19	majority	majority	NOUN
cana-4260	126	20	of	of	ADP
cana-4260	126	21	minority	minority	NOUN
cana-4260	126	22	class	class	NOUN
cana-4260	126	23	records	record	NOUN
cana-4260	126	24	and	and	CCONJ
cana-4260	126	25	need	need	VERB
cana-4260	126	26	to	to	PART
cana-4260	126	27	be	be	AUX
cana-4260	126	28	oversampled	oversample	VERB
cana-4260	126	29	.	.	PUNCT
cana-4260	127	1	this	this	DET
cana-4260	127	2	process	process	NOUN
cana-4260	127	3	makes	make	VERB
cana-4260	127	4	sure	sure	ADJ
cana-4260	127	5	oversampling	oversample	VERB
cana-4260	127	6	occurs	occur	VERB
cana-4260	127	7	in	in	ADP
cana-4260	127	8	areas	area	NOUN
cana-4260	127	9	it	it	PRON
cana-4260	127	10	is	be	AUX
cana-4260	127	11	most	most	ADV
cana-4260	127	12	beneficial	beneficial	ADJ
cana-4260	127	13	to	to	PART
cana-4260	127	14	enhance	enhance	VERB
cana-4260	127	15	class	class	NOUN
cana-4260	127	16	balance	balance	NOUN
cana-4260	127	17	.	.	PUNCT
cana-4260	128	1	clusters	cluster	NOUN
cana-4260	128	2	are	be	AUX
cana-4260	128	3	analyzed	analyze	VERB
cana-4260	128	4	by	by	ADP
cana-4260	128	5	their	their	PRON
cana-4260	128	6	imbalance	imbalance	NOUN
cana-4260	128	7	ratio	ratio	NOUN
cana-4260	128	8	,	,	PUNCT
cana-4260	128	9	which	which	PRON
cana-4260	128	10	is	be	AUX
cana-4260	128	11	calculated	calculate	VERB
cana-4260	128	12	as	as	ADP
cana-4260	128	13	:	:	PUNCT
cana-4260	128	14	imbalance	imbalance	NOUN
cana-4260	128	15	ratio	ratio	NOUN
cana-4260	128	16	=	=	NOUN
cana-4260	128	17	minoritycount(c)+1	minoritycount(c)+1	NOUN
cana-4260	128	18	/	/	SYM
cana-4260	128	19	majoritycount(c)+1	majoritycount(c)+1	NOUN
cana-4260	128	20	to	to	PART
cana-4260	128	21	calculate	calculate	VERB
cana-4260	128	22	how	how	SCONJ
cana-4260	128	23	many	many	ADJ
cana-4260	128	24	synthetic	synthetic	ADJ
cana-4260	128	25	samples	sample	NOUN
cana-4260	128	26	to	to	PART
cana-4260	128	27	produce	produce	VERB
cana-4260	128	28	per	per	ADP
cana-4260	128	29	cluster	cluster	NOUN
cana-4260	128	30	,	,	PUNCT
cana-4260	128	31	a	a	DET
cana-4260	128	32	sampling	sample	VERB
cana-4260	128	33	weight	weight	NOUN
cana-4260	128	34	is	be	AUX
cana-4260	128	35	allocated	allocate	VERB
cana-4260	128	36	to	to	ADP
cana-4260	128	37	every	every	DET
cana-4260	128	38	filtered	filter	VERB
cana-4260	128	39	cluster	cluster	NOUN
cana-4260	128	40	.	.	PUNCT
cana-4260	129	1	the	the	DET
cana-4260	129	2	weight	weight	NOUN
cana-4260	129	3	depends	depend	VERB
cana-4260	129	4	on	on	ADP
cana-4260	129	5	the	the	DET
cana-4260	129	6	minority	minority	NOUN
cana-4260	129	7	class	class	NOUN
cana-4260	129	8	density	density	NOUN
cana-4260	129	9	in	in	ADP
cana-4260	129	10	the	the	DET
cana-4260	129	11	cluster	cluster	NOUN
cana-4260	129	12	.	.	PUNCT
cana-4260	130	1	normalize	normalize	VERB
cana-4260	130	2	sparsity	sparsity	NOUN
cana-4260	130	3	values	value	NOUN
cana-4260	130	4	in	in	ADP
cana-4260	130	5	order	order	NOUN
cana-4260	130	6	to	to	PART
cana-4260	130	7	find	find	VERB
cana-4260	130	8	sampling	sample	VERB
cana-4260	130	9	weights	weight	NOUN
cana-4260	130	10	in	in	ADP
cana-4260	130	11	such	such	DET
cana-4260	130	12	a	a	DET
cana-4260	130	13	way	way	NOUN
cana-4260	130	14	that	that	PRON
cana-4260	130	15	the	the	DET
cana-4260	130	16	sum	sum	NOUN
cana-4260	130	17	of	of	ADP
cana-4260	130	18	all	all	DET
cana-4260	130	19	weights	weight	NOUN
cana-4260	130	20	equals	equal	VERB
cana-4260	130	21	1	1	NUM
cana-4260	130	22	.	.	PUNCT
cana-4260	130	23	lower	low	ADJ
cana-4260	130	24	-	-	PUNCT
cana-4260	130	25	density	density	NOUN
cana-4260	130	26	clusters	cluster	NOUN
cana-4260	130	27	(	(	PUNCT
cana-4260	130	28	more	more	ADV
cana-4260	130	29	sparsely	sparsely	ADV
cana-4260	130	30	populated	populated	ADJ
cana-4260	130	31	minority	minority	NOUN
cana-4260	130	32	points	point	NOUN
cana-4260	130	33	)	)	PUNCT
cana-4260	130	34	get	get	VERB
cana-4260	130	35	a	a	DET
cana-4260	130	36	greater	great	ADJ
cana-4260	130	37	sampling	sampling	NOUN
cana-4260	130	38	weight	weight	NOUN
cana-4260	130	39	,	,	PUNCT
cana-4260	130	40	so	so	ADV
cana-4260	130	41	more	more	ADV
cana-4260	130	42	synthetic	synthetic	ADJ
cana-4260	130	43	samples	sample	NOUN
cana-4260	130	44	are	be	AUX
cana-4260	130	45	created	create	VERB
cana-4260	130	46	within	within	ADP
cana-4260	130	47	those	those	DET
cana-4260	130	48	clusters	cluster	NOUN
cana-4260	130	49	.	.	PUNCT
cana-4260	131	1	after	after	ADP
cana-4260	131	2	the	the	DET
cana-4260	131	3	calculation	calculation	NOUN
cana-4260	131	4	of	of	ADP
cana-4260	131	5	sampling	sample	VERB
cana-4260	131	6	weights	weight	NOUN
cana-4260	131	7	,	,	PUNCT
cana-4260	131	8	smote	smote	ADJ
cana-4260	131	9	(	(	PUNCT
cana-4260	131	10	synthetic	synthetic	ADJ
cana-4260	131	11	minority	minority	NOUN
cana-4260	131	12	over	over	ADP
cana-4260	131	13	-	-	PUNCT
cana-4260	131	14	sampling	sample	VERB
cana-4260	131	15	technique	technique	NOUN
cana-4260	131	16	)	)	PUNCT
cana-4260	131	17	is	be	AUX
cana-4260	131	18	used	use	VERB
cana-4260	131	19	in	in	ADP
cana-4260	131	20	each	each	DET
cana-4260	131	21	chosen	choose	VERB
cana-4260	131	22	cluster	cluster	NOUN
cana-4260	131	23	to	to	PART
cana-4260	131	24	create	create	VERB
cana-4260	131	25	synthetic	synthetic	ADJ
cana-4260	131	26	samples	sample	NOUN
cana-4260	131	27	.	.	PUNCT
cana-4260	132	1	the	the	DET
cana-4260	132	2	number	number	NOUN
cana-4260	132	3	of	of	ADP
cana-4260	132	4	synthetic	synthetic	ADJ
cana-4260	132	5	samples	sample	NOUN
cana-4260	132	6	for	for	ADP
cana-4260	132	7	each	each	DET
cana-4260	132	8	cluster	cluster	NOUN
cana-4260	132	9	is	be	AUX
cana-4260	132	10	calculated	calculate	VERB
cana-4260	132	11	as	as	ADP
cana-4260	132	12	:	:	PUNCT
cana-4260	132	13	samplestogenerate	samplestogenerate	NOUN
cana-4260	132	14	=	=	SYM
cana-4260	132	15	samplingweight(f)×n(2	samplingweight(f)×n(2	NOUN
cana-4260	132	16	)	)	PUNCT
cana-4260	132	17	where	where	SCONJ
cana-4260	132	18	,	,	PUNCT
cana-4260	132	19	n	n	X
cana-4260	132	20	is	be	AUX
cana-4260	132	21	the	the	DET
cana-4260	132	22	number	number	NOUN
cana-4260	132	23	of	of	ADP
cana-4260	132	24	new	new	ADJ
cana-4260	132	25	samples	sample	NOUN
cana-4260	132	26	required	require	VERB
cana-4260	132	27	.	.	PUNCT
cana-4260	133	1	the	the	DET
cana-4260	133	2	smote	smote	ADJ
cana-4260	133	3	algorithm	algorithm	NOUN
cana-4260	133	4	generates	generate	VERB
cana-4260	133	5	synthetic	synthetic	ADJ
cana-4260	133	6	data	datum	NOUN
cana-4260	133	7	points	point	NOUN
cana-4260	133	8	as	as	SCONJ
cana-4260	133	9	follows	follow	VERB
cana-4260	133	10	:	:	PUNCT
cana-4260	133	11	•	•	ADV
cana-4260	133	12	choose	choose	VERB
cana-4260	133	13	a	a	DET
cana-4260	133	14	random	random	ADJ
cana-4260	133	15	minority	minority	NOUN
cana-4260	133	16	instance	instance	NOUN
cana-4260	133	17	a	a	PRON
cana-4260	133	18	in	in	ADP
cana-4260	133	19	the	the	DET
cana-4260	133	20	cluster	cluster	NOUN
cana-4260	133	21	.	.	PUNCT
cana-4260	134	1	•	•	NUM
cana-4260	134	2	identify	identify	VERB
cana-4260	134	3	its	its	PRON
cana-4260	134	4	k	k	NOUN
cana-4260	134	5	-	-	PUNCT
cana-4260	134	6	nearest	near	ADJ
cana-4260	134	7	neighbors	neighbor	NOUN
cana-4260	134	8	among	among	ADP
cana-4260	134	9	other	other	ADJ
cana-4260	134	10	minority	minority	NOUN
cana-4260	134	11	instances	instance	NOUN
cana-4260	134	12	.	.	PUNCT
cana-4260	135	1	•	•	NUM
cana-4260	135	2	choose	choose	VERB
cana-4260	135	3	one	one	NUM
cana-4260	135	4	neighbor	neighbor	NOUN
cana-4260	135	5	b	b	X
cana-4260	135	6	randomly	randomly	NOUN
cana-4260	135	7	.	.	PUNCT
cana-4260	136	1	•	•	NOUN
cana-4260	136	2	generate	generate	VERB
cana-4260	136	3	a	a	DET
cana-4260	136	4	synthetic	synthetic	ADJ
cana-4260	136	5	instance	instance	NOUN
cana-4260	136	6	by	by	ADP
cana-4260	136	7	interpolating	interpolate	VERB
cana-4260	136	8	between	between	ADP
cana-4260	136	9	a	a	PRON
cana-4260	136	10	and	and	CCONJ
cana-4260	136	11	b	b	NOUN
cana-4260	136	12	:	:	PUNCT
cana-4260	136	13	x	x	X
cana-4260	136	14	=	=	X
cana-4260	136	15	a+λ(b−a	a+λ(b−a	X
cana-4260	136	16	)	)	PUNCT
cana-4260	136	17	where	where	SCONJ
cana-4260	136	18	λ	λ	PROPN
cana-4260	136	19	is	be	AUX
cana-4260	136	20	a	a	DET
cana-4260	136	21	random	random	ADJ
cana-4260	136	22	value	value	NOUN
cana-4260	136	23	between	between	ADP
cana-4260	136	24	0	0	NUM
cana-4260	136	25	and	and	CCONJ
cana-4260	136	26	1	1	NUM
cana-4260	136	27	•	•	NOUN
cana-4260	136	28	repeat	repeat	NOUN
cana-4260	136	29	until	until	SCONJ
cana-4260	136	30	the	the	DET
cana-4260	136	31	required	require	VERB
cana-4260	136	32	number	number	NOUN
cana-4260	136	33	of	of	ADP
cana-4260	136	34	synthetic	synthetic	ADJ
cana-4260	136	35	samples	sample	NOUN
cana-4260	136	36	are	be	AUX
cana-4260	136	37	created	create	VERB
cana-4260	136	38	.	.	PUNCT
cana-4260	137	1	4	4	X
cana-4260	137	2	.	.	X
cana-4260	137	3	experimental	experimental	ADJ
cana-4260	137	4	setup	setup	NOUN
cana-4260	137	5	:	:	PUNCT
cana-4260	137	6	the	the	DET
cana-4260	137	7	experiments	experiment	NOUN
cana-4260	137	8	are	be	AUX
cana-4260	137	9	carried	carry	VERB
cana-4260	137	10	out	out	ADP
cana-4260	137	11	on	on	ADP
cana-4260	137	12	class	class	NOUN
cana-4260	137	13	-	-	PUNCT
cana-4260	137	14	imbalanced	imbalance	VERB
cana-4260	137	15	datasets	dataset	NOUN
cana-4260	137	16	to	to	PART
cana-4260	137	17	compare	compare	VERB
cana-4260	137	18	the	the	DET
cana-4260	137	19	performance	performance	NOUN
cana-4260	137	20	of	of	ADP
cana-4260	137	21	gmmsmote	gmmsmote	NOUN
cana-4260	137	22	in	in	ADP
cana-4260	137	23	dealing	deal	VERB
cana-4260	137	24	with	with	ADP
cana-4260	137	25	class	class	NOUN
cana-4260	137	26	imbalance	imbalance	NOUN
cana-4260	137	27	and	and	CCONJ
cana-4260	137	28	other	other	ADJ
cana-4260	137	29	methods	method	NOUN
cana-4260	137	30	for	for	ADP
cana-4260	137	31	class	class	NOUN
cana-4260	137	32	imbalance	imbalance	NOUN
cana-4260	137	33	handling	handling	NOUN
cana-4260	137	34	.	.	PUNCT
cana-4260	138	1	the	the	DET
cana-4260	138	2	dataset	dataset	NOUN
cana-4260	138	3	selection	selection	NOUN
cana-4260	138	4	is	be	AUX
cana-4260	138	5	on	on	ADP
cana-4260	138	6	the	the	DET
cana-4260	138	7	basis	basis	NOUN
cana-4260	138	8	of	of	ADP
cana-4260	138	9	major	major	ADJ
cana-4260	138	10	factors	factor	NOUN
cana-4260	138	11	including	include	VERB
cana-4260	138	12	class	class	NOUN
cana-4260	138	13	distribution	distribution	NOUN
cana-4260	138	14	,	,	PUNCT
cana-4260	138	15	number	number	NOUN
cana-4260	138	16	of	of	ADP
cana-4260	138	17	attributes	attribute	NOUN
cana-4260	138	18	,	,	PUNCT
cana-4260	138	19	size	size	NOUN
cana-4260	138	20	of	of	ADP
cana-4260	138	21	the	the	DET
cana-4260	138	22	dataset	dataset	NOUN
cana-4260	138	23	,	,	PUNCT
cana-4260	138	24	and	and	CCONJ
cana-4260	138	25	applicability	applicability	NOUN
cana-4260	138	26	in	in	ADP
cana-4260	138	27	real	real	ADJ
cana-4260	138	28	-	-	PUNCT
cana-4260	138	29	life	life	NOUN
cana-4260	138	30	scenarios	scenario	NOUN
cana-4260	138	31	.	.	PUNCT
cana-4260	139	1	every	every	DET
cana-4260	139	2	dataset	dataset	NOUN
cana-4260	139	3	goes	go	VERB
cana-4260	139	4	through	through	ADP
cana-4260	139	5	a	a	DET
cana-4260	139	6	preprocessing	preprocessing	NOUN
cana-4260	139	7	pipeline	pipeline	NOUN
cana-4260	139	8	for	for	ADP
cana-4260	139	9	quality	quality	NOUN
cana-4260	139	10	assurance	assurance	NOUN
cana-4260	139	11	prior	prior	ADV
cana-4260	139	12	to	to	ADP
cana-4260	139	13	the	the	DET
cana-4260	139	14	use	use	NOUN
cana-4260	139	15	of	of	ADP
cana-4260	139	16	gmmsmote	gmmsmote	NOUN
cana-4260	139	17	.	.	PUNCT
cana-4260	140	1	this	this	PRON
cana-4260	140	2	encompasses	encompass	VERB
cana-4260	140	3	dealing	deal	VERB
cana-4260	140	4	with	with	ADP
cana-4260	140	5	missing	miss	VERB
cana-4260	140	6	values	value	NOUN
cana-4260	140	7	by	by	ADP
cana-4260	140	8	applying	apply	VERB
cana-4260	140	9	methods	method	NOUN
cana-4260	140	10	such	such	ADJ
cana-4260	140	11	as	as	ADP
cana-4260	140	12	mean	mean	ADJ
cana-4260	140	13	/	/	SYM
cana-4260	140	14	mode	mode	NOUN
cana-4260	140	15	imputation	imputation	NOUN
cana-4260	140	16	or	or	CCONJ
cana-4260	140	17	predictive	predictive	ADJ
cana-4260	140	18	imputation	imputation	NOUN
cana-4260	140	19	,	,	PUNCT
cana-4260	140	20	based	base	VERB
cana-4260	140	21	on	on	ADP
cana-4260	140	22	the	the	DET
cana-4260	140	23	type	type	NOUN
cana-4260	140	24	of	of	ADP
cana-4260	140	25	missing	miss	VERB
cana-4260	140	26	data	datum	NOUN
cana-4260	140	27	.	.	PUNCT
cana-4260	141	1	numerical	numerical	ADJ
cana-4260	141	2	attributes	attribute	NOUN
cana-4260	141	3	are	be	AUX
cana-4260	141	4	normalized	normalize	VERB
cana-4260	141	5	by	by	ADP
cana-4260	141	6	employing	employ	VERB
cana-4260	141	7	min	min	ADJ
cana-4260	141	8	-	-	ADJ
cana-4260	141	9	max	max	ADJ
cana-4260	141	10	scaling	scaling	NOUN
cana-4260	141	11	or	or	CCONJ
cana-4260	141	12	standardization	standardization	NOUN
cana-4260	141	13	,	,	PUNCT
cana-4260	141	14	based	base	VERB
cana-4260	141	15	on	on	ADP
cana-4260	141	16	the	the	DET
cana-4260	141	17	need	need	NOUN
cana-4260	141	18	of	of	ADP
cana-4260	141	19	the	the	DET
cana-4260	141	20	classifier	classifier	NOUN
cana-4260	141	21	,	,	PUNCT
cana-4260	141	22	in	in	ADP
cana-4260	141	23	order	order	NOUN
cana-4260	141	24	to	to	PART
cana-4260	141	25	avoid	avoid	VERB
cana-4260	141	26	feature	feature	NOUN
cana-4260	141	27	communications	communication	NOUN
cana-4260	141	28	on	on	ADP
cana-4260	141	29	applied	apply	VERB
cana-4260	141	30	nonlinear	nonlinear	ADJ
cana-4260	141	31	analysis	analysis	NOUN
cana-4260	141	32	issn	issn	NOUN
cana-4260	141	33	:	:	PUNCT
cana-4260	141	34	1074	1074	NUM
cana-4260	141	35	-	-	PUNCT
cana-4260	141	36	133x	133x	NUM
cana-4260	141	37	vol	vol	NOUN
cana-4260	141	38	32	32	NUM
cana-4260	142	1	no	no	NOUN
cana-4260	142	2	.	.	PUNCT
cana-4260	143	1	9s	9s	NUM
cana-4260	143	2	(	(	PUNCT
cana-4260	143	3	2025	2025	NUM
cana-4260	143	4	)	)	PUNCT
cana-4260	143	5	1637	1637	NUM
cana-4260	143	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-4260	143	7	magnitude	magnitude	NOUN
cana-4260	143	8	creating	create	VERB
cana-4260	143	9	biases	bias	NOUN
cana-4260	143	10	.	.	PUNCT
cana-4260	144	1	categorical	categorical	ADJ
cana-4260	144	2	features	feature	NOUN
cana-4260	144	3	are	be	AUX
cana-4260	144	4	converted	convert	VERB
cana-4260	144	5	through	through	ADP
cana-4260	144	6	one	one	NUM
cana-4260	144	7	-	-	PUNCT
cana-4260	144	8	hot	hot	ADJ
cana-4260	144	9	encoding	encoding	NOUN
cana-4260	144	10	or	or	CCONJ
cana-4260	144	11	label	label	NOUN
cana-4260	144	12	encoding	encoding	NOUN
cana-4260	144	13	for	for	ADP
cana-4260	144	14	machine	machine	NOUN
cana-4260	144	15	learning	learning	NOUN
cana-4260	144	16	model	model	NOUN
cana-4260	144	17	compatibility	compatibility	NOUN
cana-4260	144	18	.	.	PUNCT
cana-4260	145	1	data	datum	NOUN
cana-4260	145	2	with	with	ADP
cana-4260	145	3	extreme	extreme	ADJ
cana-4260	145	4	imbalance	imbalance	NOUN
cana-4260	145	5	is	be	AUX
cana-4260	145	6	also	also	ADV
cana-4260	145	7	subject	subject	ADJ
cana-4260	145	8	to	to	ADP
cana-4260	145	9	initial	initial	ADJ
cana-4260	145	10	exploratory	exploratory	ADJ
cana-4260	145	11	analysis	analysis	NOUN
cana-4260	145	12	where	where	SCONJ
cana-4260	145	13	class	class	NOUN
cana-4260	145	14	distributions	distribution	NOUN
cana-4260	145	15	,	,	PUNCT
cana-4260	145	16	feature	feature	NOUN
cana-4260	145	17	correlations	correlation	NOUN
cana-4260	145	18	,	,	PUNCT
cana-4260	145	19	and	and	CCONJ
cana-4260	145	20	biases	bias	NOUN
cana-4260	145	21	are	be	AUX
cana-4260	145	22	explored	explore	VERB
cana-4260	145	23	to	to	PART
cana-4260	145	24	know	know	VERB
cana-4260	145	25	the	the	DET
cana-4260	145	26	characteristics	characteristic	NOUN
cana-4260	145	27	of	of	ADP
cana-4260	145	28	the	the	DET
cana-4260	145	29	data	datum	NOUN
cana-4260	145	30	prior	prior	ADV
cana-4260	145	31	to	to	ADP
cana-4260	145	32	resampling	resample	VERB
cana-4260	145	33	.	.	PUNCT
cana-4260	146	1	5	5	X
cana-4260	146	2	.	.	X
cana-4260	146	3	computation	computation	NOUN
cana-4260	146	4	environment	environment	NOUN
cana-4260	146	5	:	:	PUNCT
cana-4260	146	6	experiments	experiment	NOUN
cana-4260	146	7	are	be	AUX
cana-4260	146	8	conducted	conduct	VERB
cana-4260	146	9	on	on	ADP
cana-4260	146	10	a	a	DET
cana-4260	146	11	high	high	ADJ
cana-4260	146	12	-	-	PUNCT
cana-4260	146	13	end	end	NOUN
cana-4260	146	14	computing	compute	VERB
cana-4260	146	15	environment	environment	NOUN
cana-4260	146	16	using	use	VERB
cana-4260	146	17	python	python	NOUN
cana-4260	146	18	-	-	PUNCT
cana-4260	146	19	based	base	VERB
cana-4260	146	20	development	development	NOUN
cana-4260	146	21	.	.	PUNCT
cana-4260	147	1	the	the	DET
cana-4260	147	2	prominent	prominent	ADJ
cana-4260	147	3	libraries	library	NOUN
cana-4260	147	4	are	be	AUX
cana-4260	147	5	scikit	scikit	NOUN
cana-4260	147	6	-	-	PUNCT
cana-4260	147	7	learn	learn	NOUN
cana-4260	147	8	,	,	PUNCT
cana-4260	147	9	numpy	numpy	NOUN
cana-4260	147	10	,	,	PUNCT
cana-4260	147	11	pandas	panda	NOUN
cana-4260	147	12	,	,	PUNCT
cana-4260	147	13	tensorflow	tensorflow	NOUN
cana-4260	147	14	/	/	SYM
cana-4260	147	15	pytorch	pytorch	NOUN
cana-4260	147	16	(	(	PUNCT
cana-4260	147	17	in	in	ADP
cana-4260	147	18	case	case	NOUN
cana-4260	147	19	deep	deep	ADJ
cana-4260	147	20	models	model	NOUN
cana-4260	147	21	are	be	AUX
cana-4260	147	22	involved	involve	VERB
cana-4260	147	23	)	)	PUNCT
cana-4260	147	24	,	,	PUNCT
cana-4260	147	25	and	and	CCONJ
cana-4260	147	26	imbalanced	imbalance	VERB
cana-4260	147	27	-	-	PUNCT
cana-4260	147	28	learn	learn	VERB
cana-4260	147	29	for	for	ADP
cana-4260	147	30	resampling	resample	VERB
cana-4260	147	31	methods	method	NOUN
cana-4260	147	32	.	.	PUNCT
cana-4260	148	1	the	the	DET
cana-4260	148	2	hardware	hardware	NOUN
cana-4260	148	3	configuration	configuration	NOUN
cana-4260	148	4	comprises	comprise	VERB
cana-4260	148	5	a	a	DET
cana-4260	148	6	multi	multi	ADJ
cana-4260	148	7	-	-	ADJ
cana-4260	148	8	core	core	ADJ
cana-4260	148	9	cpu	cpu	NOUN
cana-4260	148	10	,	,	PUNCT
cana-4260	148	11	gpu	gpu	NOUN
cana-4260	148	12	support	support	NOUN
cana-4260	148	13	(	(	PUNCT
cana-4260	148	14	in	in	ADP
cana-4260	148	15	case	case	NOUN
cana-4260	148	16	deep	deep	ADJ
cana-4260	148	17	models	model	NOUN
cana-4260	148	18	are	be	AUX
cana-4260	148	19	utilized	utilize	VERB
cana-4260	148	20	)	)	PUNCT
cana-4260	148	21	,	,	PUNCT
cana-4260	148	22	and	and	CCONJ
cana-4260	148	23	ample	ample	ADJ
cana-4260	148	24	ram	ram	NOUN
cana-4260	148	25	for	for	ADP
cana-4260	148	26	efficient	efficient	ADJ
cana-4260	148	27	execution	execution	NOUN
cana-4260	148	28	.	.	PUNCT
cana-4260	149	1	6	6	X
cana-4260	149	2	.	.	X
cana-4260	149	3	metrics	metric	NOUN
cana-4260	149	4	:	:	PUNCT
cana-4260	149	5	in	in	ADP
cana-4260	149	6	order	order	NOUN
cana-4260	149	7	to	to	PART
cana-4260	149	8	determine	determine	VERB
cana-4260	149	9	the	the	DET
cana-4260	149	10	performance	performance	NOUN
cana-4260	149	11	of	of	ADP
cana-4260	149	12	gmm	gmm	NOUN
cana-4260	149	13	-	-	PUNCT
cana-4260	149	14	smote	smote	ADJ
cana-4260	149	15	,	,	PUNCT
cana-4260	149	16	some	some	DET
cana-4260	149	17	evaluation	evaluation	NOUN
cana-4260	149	18	measures	measure	NOUN
cana-4260	149	19	are	be	AUX
cana-4260	149	20	taken	take	VERB
cana-4260	149	21	into	into	ADP
cana-4260	149	22	consideration	consideration	NOUN
cana-4260	149	23	to	to	PART
cana-4260	149	24	quantify	quantify	VERB
cana-4260	149	25	its	its	PRON
cana-4260	149	26	effect	effect	NOUN
cana-4260	149	27	on	on	ADP
cana-4260	149	28	unbalanced	unbalanced	ADJ
cana-4260	149	29	datasets	dataset	NOUN
cana-4260	149	30	.	.	PUNCT
cana-4260	150	1	one	one	NUM
cana-4260	150	2	of	of	ADP
cana-4260	150	3	the	the	DET
cana-4260	150	4	most	most	ADV
cana-4260	150	5	important	important	ADJ
cana-4260	150	6	parameters	parameter	NOUN
cana-4260	150	7	is	be	AUX
cana-4260	150	8	the	the	DET
cana-4260	150	9	value	value	NOUN
cana-4260	150	10	of	of	ADP
cana-4260	150	11	k	k	NOUN
cana-4260	150	12	,	,	PUNCT
cana-4260	150	13	which	which	PRON
cana-4260	150	14	is	be	AUX
cana-4260	150	15	the	the	DET
cana-4260	150	16	number	number	NOUN
cana-4260	150	17	of	of	ADP
cana-4260	150	18	clusters	cluster	NOUN
cana-4260	150	19	created	create	VERB
cana-4260	150	20	by	by	ADP
cana-4260	150	21	the	the	DET
cana-4260	150	22	gaussian	gaussian	ADJ
cana-4260	150	23	mixture	mixture	NOUN
cana-4260	150	24	model	model	NOUN
cana-4260	150	25	(	(	PUNCT
cana-4260	150	26	gmm	gmm	PROPN
cana-4260	150	27	)	)	PUNCT
cana-4260	150	28	.	.	PUNCT
cana-4260	151	1	the	the	DET
cana-4260	151	2	value	value	NOUN
cana-4260	151	3	of	of	ADP
cana-4260	151	4	k	k	PROPN
cana-4260	151	5	has	have	VERB
cana-4260	151	6	a	a	DET
cana-4260	151	7	deep	deep	ADJ
cana-4260	151	8	impact	impact	NOUN
cana-4260	151	9	on	on	ADP
cana-4260	151	10	the	the	DET
cana-4260	151	11	quality	quality	NOUN
cana-4260	151	12	of	of	ADP
cana-4260	151	13	generated	generate	VERB
cana-4260	151	14	synthetic	synthetic	ADJ
cana-4260	151	15	samples	sample	NOUN
cana-4260	151	16	.	.	PUNCT
cana-4260	152	1	a	a	DET
cana-4260	152	2	smaller	small	ADJ
cana-4260	152	3	value	value	NOUN
cana-4260	152	4	of	of	ADP
cana-4260	152	5	k	k	PROPN
cana-4260	152	6	could	could	AUX
cana-4260	152	7	result	result	VERB
cana-4260	152	8	in	in	ADP
cana-4260	152	9	overly	overly	ADV
cana-4260	152	10	simplified	simplified	ADJ
cana-4260	152	11	clusters	cluster	NOUN
cana-4260	152	12	that	that	PRON
cana-4260	152	13	do	do	AUX
cana-4260	152	14	not	not	PART
cana-4260	152	15	pick	pick	VERB
cana-4260	152	16	up	up	ADP
cana-4260	152	17	the	the	DET
cana-4260	152	18	intricacy	intricacy	NOUN
cana-4260	152	19	of	of	ADP
cana-4260	152	20	minority	minority	NOUN
cana-4260	152	21	class	class	NOUN
cana-4260	152	22	distributions	distribution	NOUN
cana-4260	152	23	,	,	PUNCT
cana-4260	152	24	while	while	SCONJ
cana-4260	152	25	an	an	DET
cana-4260	152	26	increased	increase	VERB
cana-4260	152	27	value	value	NOUN
cana-4260	152	28	of	of	ADP
cana-4260	152	29	k	k	PROPN
cana-4260	152	30	could	could	AUX
cana-4260	152	31	produce	produce	VERB
cana-4260	152	32	overly	overly	ADV
cana-4260	152	33	localized	localize	VERB
cana-4260	152	34	clusters	cluster	NOUN
cana-4260	152	35	and	and	CCONJ
cana-4260	152	36	thus	thus	ADV
cana-4260	152	37	restrict	restrict	VERB
cana-4260	152	38	the	the	DET
cana-4260	152	39	model	model	NOUN
cana-4260	152	40	's	's	PART
cana-4260	152	41	generalization	generalization	NOUN
cana-4260	152	42	capability	capability	NOUN
cana-4260	152	43	.	.	PUNCT
cana-4260	153	1	hence	hence	ADV
cana-4260	153	2	,	,	PUNCT
cana-4260	153	3	an	an	DET
cana-4260	153	4	optimal	optimal	ADJ
cana-4260	153	5	value	value	NOUN
cana-4260	153	6	of	of	ADP
cana-4260	153	7	k	k	PROPN
cana-4260	153	8	needs	need	VERB
cana-4260	153	9	to	to	PART
cana-4260	153	10	be	be	AUX
cana-4260	153	11	chosen	choose	VERB
cana-4260	153	12	so	so	SCONJ
cana-4260	153	13	that	that	SCONJ
cana-4260	153	14	the	the	DET
cana-4260	153	15	synthesized	synthesize	VERB
cana-4260	153	16	data	data	NOUN
cana-4260	153	17	is	be	AUX
cana-4260	153	18	able	able	ADJ
cana-4260	153	19	to	to	PART
cana-4260	153	20	optimize	optimize	VERB
cana-4260	153	21	model	model	NOUN
cana-4260	153	22	learning	learning	NOUN
cana-4260	153	23	.	.	PUNCT
cana-4260	154	1	another	another	DET
cana-4260	154	2	significant	significant	ADJ
cana-4260	154	3	measure	measure	NOUN
cana-4260	154	4	is	be	AUX
cana-4260	154	5	accuracy	accuracy	NOUN
cana-4260	154	6	,	,	PUNCT
cana-4260	154	7	which	which	PRON
cana-4260	154	8	indicates	indicate	VERB
cana-4260	154	9	the	the	DET
cana-4260	154	10	overall	overall	ADJ
cana-4260	154	11	accuracy	accuracy	NOUN
cana-4260	154	12	of	of	ADP
cana-4260	154	13	the	the	DET
cana-4260	154	14	classifier	classifier	NOUN
cana-4260	154	15	's	's	PART
cana-4260	154	16	predictions	prediction	NOUN
cana-4260	154	17	.	.	PUNCT
cana-4260	155	1	in	in	ADP
cana-4260	155	2	imbalanced	imbalanced	ADJ
cana-4260	155	3	datasets	dataset	NOUN
cana-4260	155	4	,	,	PUNCT
cana-4260	155	5	however	however	ADV
cana-4260	155	6	,	,	PUNCT
cana-4260	155	7	accuracy	accuracy	NOUN
cana-4260	155	8	is	be	AUX
cana-4260	155	9	deceptive	deceptive	ADJ
cana-4260	155	10	since	since	SCONJ
cana-4260	155	11	a	a	DET
cana-4260	155	12	model	model	NOUN
cana-4260	155	13	may	may	AUX
cana-4260	155	14	obtain	obtain	VERB
cana-4260	155	15	high	high	ADJ
cana-4260	155	16	accuracy	accuracy	NOUN
cana-4260	155	17	through	through	ADP
cana-4260	155	18	simply	simply	ADV
cana-4260	155	19	predicting	predict	VERB
cana-4260	155	20	the	the	DET
cana-4260	155	21	majority	majority	NOUN
cana-4260	155	22	class	class	NOUN
cana-4260	155	23	more	more	ADV
cana-4260	155	24	frequently	frequently	ADV
cana-4260	155	25	and	and	CCONJ
cana-4260	155	26	ignoring	ignore	VERB
cana-4260	155	27	the	the	DET
cana-4260	155	28	minority	minority	NOUN
cana-4260	155	29	class	class	NOUN
cana-4260	155	30	.	.	PUNCT
cana-4260	156	1	because	because	SCONJ
cana-4260	156	2	of	of	ADP
cana-4260	156	3	this	this	DET
cana-4260	156	4	drawback	drawback	NOUN
cana-4260	156	5	,	,	PUNCT
cana-4260	156	6	it	it	PRON
cana-4260	156	7	is	be	AUX
cana-4260	156	8	useful	useful	ADJ
cana-4260	156	9	to	to	PART
cana-4260	156	10	examine	examine	VERB
cana-4260	156	11	other	other	ADJ
cana-4260	156	12	metrics	metric	NOUN
cana-4260	156	13	that	that	PRON
cana-4260	156	14	offer	offer	VERB
cana-4260	156	15	most	most	ADV
cana-4260	156	16	balanced	balanced	ADJ
cana-4260	156	17	measure	measure	NOUN
cana-4260	156	18	of	of	ADP
cana-4260	156	19	the	the	DET
cana-4260	156	20	performance	performance	NOUN
cana-4260	156	21	of	of	ADP
cana-4260	156	22	classification	classification	NOUN
cana-4260	156	23	.	.	PUNCT
cana-4260	157	1	the	the	DET
cana-4260	157	2	auc	auc	NOUN
cana-4260	157	3	-	-	PUNCT
cana-4260	157	4	roc	roc	NOUN
cana-4260	157	5	score	score	NOUN
cana-4260	157	6	is	be	AUX
cana-4260	157	7	especially	especially	ADV
cana-4260	157	8	helpful	helpful	ADJ
cana-4260	157	9	in	in	ADP
cana-4260	157	10	measuring	measure	VERB
cana-4260	157	11	how	how	SCONJ
cana-4260	157	12	well	well	ADV
cana-4260	157	13	the	the	DET
cana-4260	157	14	model	model	NOUN
cana-4260	157	15	separates	separate	VERB
cana-4260	157	16	various	various	ADJ
cana-4260	157	17	classes	class	NOUN
cana-4260	157	18	.	.	PUNCT
cana-4260	158	1	the	the	PRON
cana-4260	158	2	higher	high	ADJ
cana-4260	158	3	the	the	DET
cana-4260	158	4	auc	auc	NOUN
cana-4260	158	5	-	-	PUNCT
cana-4260	158	6	roc	roc	NOUN
cana-4260	158	7	,	,	PUNCT
cana-4260	158	8	the	the	PRON
cana-4260	158	9	better	well	ADJ
cana-4260	158	10	the	the	DET
cana-4260	158	11	trade	trade	NOUN
cana-4260	158	12	-	-	PUNCT
cana-4260	158	13	off	off	NOUN
cana-4260	158	14	between	between	ADP
cana-4260	158	15	sensitivity	sensitivity	NOUN
cana-4260	158	16	(	(	PUNCT
cana-4260	158	17	recall	recall	NOUN
cana-4260	158	18	)	)	PUNCT
cana-4260	158	19	and	and	CCONJ
cana-4260	158	20	specificity	specificity	NOUN
cana-4260	158	21	,	,	PUNCT
cana-4260	158	22	and	and	CCONJ
cana-4260	158	23	thus	thus	ADV
cana-4260	158	24	it	it	PRON
cana-4260	158	25	is	be	AUX
cana-4260	158	26	a	a	DET
cana-4260	158	27	good	good	ADJ
cana-4260	158	28	indicator	indicator	NOUN
cana-4260	158	29	for	for	ADP
cana-4260	158	30	measurement	measurement	NOUN
cana-4260	158	31	of	of	ADP
cana-4260	158	32	the	the	DET
cana-4260	158	33	effect	effect	NOUN
cana-4260	158	34	of	of	ADP
cana-4260	158	35	gmm	gmm	NOUN
cana-4260	158	36	-	-	PUNCT
cana-4260	158	37	smote	smote	ADJ
cana-4260	158	38	.	.	PUNCT
cana-4260	159	1	by	by	ADP
cana-4260	159	2	enhancing	enhance	VERB
cana-4260	159	3	the	the	DET
cana-4260	159	4	minority	minority	NOUN
cana-4260	159	5	class	class	NOUN
cana-4260	159	6	representation	representation	NOUN
cana-4260	159	7	,	,	PUNCT
cana-4260	159	8	gmm	gmm	NOUN
cana-4260	159	9	-	-	PUNCT
cana-4260	159	10	smote	smote	ADJ
cana-4260	159	11	seeks	seek	VERB
cana-4260	159	12	to	to	PART
cana-4260	159	13	increase	increase	VERB
cana-4260	159	14	auc	auc	ADJ
cana-4260	159	15	-	-	PUNCT
cana-4260	159	16	roc	roc	NOUN
cana-4260	159	17	scores	score	NOUN
cana-4260	159	18	so	so	SCONJ
cana-4260	159	19	that	that	SCONJ
cana-4260	159	20	the	the	DET
cana-4260	159	21	model	model	NOUN
cana-4260	159	22	not	not	PART
cana-4260	159	23	only	only	ADV
cana-4260	159	24	prefers	prefer	VERB
cana-4260	159	25	the	the	DET
cana-4260	159	26	majority	majority	NOUN
cana-4260	159	27	class	class	NOUN
cana-4260	159	28	but	but	CCONJ
cana-4260	159	29	learns	learn	VERB
cana-4260	159	30	to	to	PART
cana-4260	159	31	distinguish	distinguish	VERB
cana-4260	159	32	between	between	ADP
cana-4260	159	33	both	both	DET
cana-4260	159	34	classes	class	NOUN
cana-4260	159	35	effectively	effectively	ADV
cana-4260	159	36	.	.	PUNCT
cana-4260	160	1	these	these	DET
cana-4260	160	2	metrics	metric	NOUN
cana-4260	160	3	collectively	collectively	ADV
cana-4260	160	4	offer	offer	VERB
cana-4260	160	5	a	a	DET
cana-4260	160	6	complete	complete	ADJ
cana-4260	160	7	assessment	assessment	NOUN
cana-4260	160	8	of	of	ADP
cana-4260	160	9	gmm	gmm	NOUN
cana-4260	160	10	-	-	PUNCT
cana-4260	160	11	smote	smote	NOUN
cana-4260	160	12	's	's	PART
cana-4260	160	13	contribution	contribution	NOUN
cana-4260	160	14	towards	towards	ADP
cana-4260	160	15	improving	improve	VERB
cana-4260	160	16	classification	classification	NOUN
cana-4260	160	17	performance	performance	NOUN
cana-4260	160	18	on	on	ADP
cana-4260	160	19	imbalanced	imbalanced	ADJ
cana-4260	160	20	datasets	dataset	NOUN
cana-4260	160	21	.	.	PUNCT
cana-4260	161	1	7	7	X
cana-4260	161	2	.	.	X
cana-4260	161	3	results	result	VERB
cana-4260	161	4	the	the	DET
cana-4260	161	5	findings	finding	NOUN
cana-4260	161	6	offer	offer	VERB
cana-4260	161	7	the	the	DET
cana-4260	161	8	performance	performance	NOUN
cana-4260	161	9	of	of	ADP
cana-4260	161	10	gmm	gmm	NOUN
cana-4260	161	11	-	-	PUNCT
cana-4260	161	12	smote	smote	ADJ
cana-4260	161	13	in	in	ADP
cana-4260	161	14	contrast	contrast	NOUN
cana-4260	161	15	to	to	ADP
cana-4260	161	16	other	other	ADJ
cana-4260	161	17	oversampling	oversampling	ADJ
cana-4260	161	18	methods	method	NOUN
cana-4260	161	19	,	,	PUNCT
cana-4260	161	20	like	like	ADP
cana-4260	161	21	kmeanssmote	kmeanssmote	NOUN
cana-4260	161	22	,	,	PUNCT
cana-4260	161	23	kmeans	kmean	NOUN
cana-4260	161	24	-	-	PUNCT
cana-4260	161	25	adasyn	adasyn	NOUN
cana-4260	161	26	,	,	PUNCT
cana-4260	161	27	and	and	CCONJ
cana-4260	161	28	gmm	gmm	NOUN
cana-4260	161	29	-	-	PUNCT
cana-4260	161	30	adasyn	adasyn	PROPN
cana-4260	161	31	,	,	PUNCT
cana-4260	161	32	over	over	ADP
cana-4260	161	33	various	various	ADJ
cana-4260	161	34	datasets	dataset	NOUN
cana-4260	161	35	.	.	PUNCT
cana-4260	162	1	the	the	DET
cana-4260	162	2	comparison	comparison	NOUN
cana-4260	162	3	takes	take	VERB
cana-4260	162	4	into	into	ADP
cana-4260	162	5	account	account	NOUN
cana-4260	162	6	primary	primary	ADJ
cana-4260	162	7	measures	measure	NOUN
cana-4260	162	8	,	,	PUNCT
cana-4260	162	9	like	like	ADP
cana-4260	162	10	accuracy	accuracy	NOUN
cana-4260	162	11	,	,	PUNCT
cana-4260	162	12	auc	auc	NOUN
cana-4260	162	13	-	-	PUNCT
cana-4260	162	14	roc	roc	NOUN
cana-4260	162	15	score	score	NOUN
cana-4260	162	16	,	,	PUNCT
cana-4260	162	17	and	and	CCONJ
cana-4260	162	18	k	k	PROPN
cana-4260	162	19	value	value	NOUN
cana-4260	162	20	,	,	PUNCT
cana-4260	162	21	to	to	PART
cana-4260	162	22	gauge	gauge	VERB
cana-4260	162	23	the	the	DET
cana-4260	162	24	performance	performance	NOUN
cana-4260	162	25	of	of	ADP
cana-4260	162	26	the	the	DET
cana-4260	162	27	methods	method	NOUN
cana-4260	162	28	in	in	ADP
cana-4260	162	29	managing	manage	VERB
cana-4260	162	30	class	class	NOUN
cana-4260	162	31	imbalance	imbalance	NOUN
cana-4260	162	32	.	.	PUNCT
cana-4260	163	1	the	the	DET
cana-4260	163	2	experiments	experiment	NOUN
cana-4260	163	3	are	be	AUX
cana-4260	163	4	performed	perform	VERB
cana-4260	163	5	on	on	ADP
cana-4260	163	6	various	various	ADJ
cana-4260	163	7	datasets	dataset	NOUN
cana-4260	163	8	like	like	ADP
cana-4260	163	9	breast	breast	NOUN
cana-4260	163	10	cancer	cancer	NOUN
cana-4260	163	11	,	,	PUNCT
cana-4260	163	12	crx	crx	PROPN
cana-4260	163	13	,	,	PUNCT
cana-4260	163	14	and	and	CCONJ
cana-4260	163	15	churn	churn	VERB
cana-4260	163	16	bigml	bigml	NOUN
cana-4260	163	17	with	with	ADP
cana-4260	163	18	classifiers	classifier	NOUN
cana-4260	163	19	like	like	ADP
cana-4260	163	20	random	random	ADJ
cana-4260	163	21	forest	forest	NOUN
cana-4260	163	22	,	,	PUNCT
cana-4260	163	23	svm	svm	ADJ
cana-4260	163	24	,	,	PUNCT
cana-4260	163	25	logistic	logistic	ADJ
cana-4260	163	26	regression	regression	NOUN
cana-4260	163	27	,	,	PUNCT
cana-4260	163	28	and	and	CCONJ
cana-4260	163	29	neural	neural	ADJ
cana-4260	163	30	networks	network	NOUN
cana-4260	163	31	.	.	PUNCT
cana-4260	164	1	the	the	DET
cana-4260	164	2	results	result	NOUN
cana-4260	164	3	are	be	AUX
cana-4260	164	4	presented	present	VERB
cana-4260	164	5	in	in	ADP
cana-4260	164	6	tables	table	NOUN
cana-4260	164	7	as	as	ADV
cana-4260	164	8	well	well	ADV
cana-4260	164	9	as	as	ADP
cana-4260	164	10	through	through	ADP
cana-4260	164	11	auc	auc	NOUN
cana-4260	164	12	-	-	PUNCT
cana-4260	164	13	roc	roc	NOUN
cana-4260	164	14	curves	curve	NOUN
cana-4260	164	15	,	,	PUNCT
cana-4260	164	16	emphasizing	emphasize	VERB
cana-4260	164	17	the	the	DET
cana-4260	164	18	effect	effect	NOUN
cana-4260	164	19	of	of	ADP
cana-4260	164	20	various	various	ADJ
cana-4260	164	21	resampling	resample	VERB
cana-4260	164	22	methods	method	NOUN
cana-4260	164	23	.	.	PUNCT
cana-4260	165	1	communications	communication	NOUN
cana-4260	165	2	on	on	ADP
cana-4260	165	3	applied	apply	VERB
cana-4260	165	4	nonlinear	nonlinear	ADJ
cana-4260	165	5	analysis	analysis	NOUN
cana-4260	165	6	issn	issn	NOUN
cana-4260	165	7	:	:	PUNCT
cana-4260	165	8	1074	1074	NUM
cana-4260	165	9	-	-	PUNCT
cana-4260	165	10	133x	133x	NUM
cana-4260	165	11	vol	vol	NOUN
cana-4260	165	12	32	32	NUM
cana-4260	165	13	no	no	NOUN
cana-4260	165	14	.	.	PUNCT
cana-4260	166	1	9s	9s	NUM
cana-4260	166	2	(	(	PUNCT
cana-4260	166	3	2025	2025	NUM
cana-4260	166	4	)	)	PUNCT
cana-4260	166	5	1638	1638	NUM
cana-4260	166	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-4260	167	1	the	the	DET
cana-4260	167	2	comparison	comparison	NOUN
cana-4260	167	3	also	also	ADV
cana-4260	167	4	involves	involve	VERB
cana-4260	167	5	a	a	DET
cana-4260	167	6	training	training	NOUN
cana-4260	167	7	time	time	NOUN
cana-4260	167	8	analysis	analysis	NOUN
cana-4260	167	9	to	to	PART
cana-4260	167	10	assess	assess	VERB
cana-4260	167	11	computational	computational	ADJ
cana-4260	167	12	efficiency	efficiency	NOUN
cana-4260	167	13	.	.	PUNCT
cana-4260	168	1	the	the	DET
cana-4260	168	2	results	result	NOUN
cana-4260	168	3	show	show	VERB
cana-4260	168	4	that	that	SCONJ
cana-4260	168	5	gmm	gmm	NOUN
cana-4260	168	6	-	-	PUNCT
cana-4260	168	7	smote	smote	ADJ
cana-4260	168	8	efficiently	efficiently	ADV
cana-4260	168	9	enhances	enhance	VERB
cana-4260	168	10	classification	classification	NOUN
cana-4260	168	11	performance	performance	NOUN
cana-4260	168	12	with	with	ADP
cana-4260	168	13	an	an	DET
cana-4260	168	14	even	even	ADJ
cana-4260	168	15	decision	decision	NOUN
cana-4260	168	16	boundary	boundary	NOUN
cana-4260	168	17	,	,	PUNCT
cana-4260	168	18	proving	prove	VERB
cana-4260	168	19	its	its	PRON
cana-4260	168	20	practical	practical	ADJ
cana-4260	168	21	benefit	benefit	NOUN
cana-4260	168	22	in	in	ADP
cana-4260	168	23	imbalanced	imbalanced	ADJ
cana-4260	168	24	learning	learning	NOUN
cana-4260	168	25	tasks	task	NOUN
cana-4260	168	26	.	.	PUNCT
cana-4260	169	1	8	8	X
cana-4260	169	2	.	.	X
cana-4260	169	3	discussion	discussion	NOUN
cana-4260	169	4	the	the	DET
cana-4260	169	5	motivation	motivation	NOUN
cana-4260	169	6	behind	behind	ADP
cana-4260	169	7	this	this	DET
cana-4260	169	8	study	study	NOUN
cana-4260	169	9	is	be	AUX
cana-4260	169	10	to	to	PART
cana-4260	169	11	address	address	VERB
cana-4260	169	12	the	the	DET
cana-4260	169	13	limitations	limitation	NOUN
cana-4260	169	14	of	of	ADP
cana-4260	169	15	existing	exist	VERB
cana-4260	169	16	oversampling	oversample	VERB
cana-4260	169	17	techniques	technique	NOUN
cana-4260	169	18	by	by	ADP
cana-4260	169	19	leveraging	leverage	VERB
cana-4260	169	20	gaussian	gaussian	ADJ
cana-4260	169	21	mixture	mixture	NOUN
cana-4260	169	22	models	model	NOUN
cana-4260	169	23	(	(	PUNCT
cana-4260	169	24	gmm	gmm	NOUN
cana-4260	169	25	)	)	PUNCT
cana-4260	169	26	for	for	ADP
cana-4260	169	27	clustering	cluster	VERB
cana-4260	169	28	-	-	PUNCT
cana-4260	169	29	based	base	VERB
cana-4260	169	30	synthetic	synthetic	ADJ
cana-4260	169	31	data	data	NOUN
cana-4260	169	32	generation	generation	NOUN
cana-4260	169	33	.	.	PUNCT
cana-4260	170	1	standard	standard	ADJ
cana-4260	170	2	methods	method	NOUN
cana-4260	170	3	like	like	ADP
cana-4260	170	4	smote	smote	NOUN
cana-4260	170	5	create	create	VERB
cana-4260	170	6	synthetic	synthetic	ADJ
cana-4260	170	7	samples	sample	NOUN
cana-4260	170	8	by	by	ADP
cana-4260	170	9	linear	linear	ADJ
cana-4260	170	10	interpolation	interpolation	NOUN
cana-4260	170	11	,	,	PUNCT
cana-4260	170	12	which	which	PRON
cana-4260	170	13	can	can	AUX
cana-4260	170	14	sometimes	sometimes	ADV
cana-4260	170	15	introduce	introduce	VERB
cana-4260	170	16	noisy	noisy	ADJ
cana-4260	170	17	or	or	CCONJ
cana-4260	170	18	unrealistic	unrealistic	ADJ
cana-4260	170	19	samples	sample	NOUN
cana-4260	170	20	,	,	PUNCT
cana-4260	170	21	especially	especially	ADV
cana-4260	170	22	in	in	ADP
cana-4260	170	23	complex	complex	ADJ
cana-4260	170	24	feature	feature	NOUN
cana-4260	170	25	spaces	space	NOUN
cana-4260	170	26	.	.	PUNCT
cana-4260	171	1	gmmsmote	gmmsmote	NOUN
cana-4260	171	2	improves	improve	VERB
cana-4260	171	3	upon	upon	SCONJ
cana-4260	171	4	this	this	PRON
cana-4260	171	5	by	by	ADP
cana-4260	171	6	using	use	VERB
cana-4260	171	7	probabilistic	probabilistic	ADJ
cana-4260	171	8	modeling	modeling	NOUN
cana-4260	171	9	to	to	PART
cana-4260	171	10	identify	identify	VERB
cana-4260	171	11	underrepresented	underrepresented	ADJ
cana-4260	171	12	clusters	cluster	NOUN
cana-4260	171	13	and	and	CCONJ
cana-4260	171	14	generate	generate	VERB
cana-4260	171	15	synthetic	synthetic	ADJ
cana-4260	171	16	data	datum	NOUN
cana-4260	171	17	that	that	PRON
cana-4260	171	18	aligns	align	VERB
cana-4260	171	19	more	more	ADV
cana-4260	171	20	naturally	naturally	ADV
cana-4260	171	21	with	with	ADP
cana-4260	171	22	the	the	DET
cana-4260	171	23	distribution	distribution	NOUN
cana-4260	171	24	of	of	ADP
cana-4260	171	25	the	the	DET
cana-4260	171	26	minority	minority	NOUN
cana-4260	171	27	class	class	NOUN
cana-4260	171	28	.	.	PUNCT
cana-4260	172	1	through	through	ADP
cana-4260	172	2	our	our	PRON
cana-4260	172	3	experiments	experiment	NOUN
cana-4260	172	4	,	,	PUNCT
cana-4260	172	5	we	we	PRON
cana-4260	172	6	successfully	successfully	ADV
cana-4260	172	7	demonstrated	demonstrate	VERB
cana-4260	172	8	that	that	SCONJ
cana-4260	172	9	:	:	PUNCT
cana-4260	172	10	•	•	NUM
cana-4260	172	11	gmm	gmm	NOUN
cana-4260	172	12	-	-	PUNCT
cana-4260	172	13	smote	smote	ADJ
cana-4260	172	14	enhances	enhance	VERB
cana-4260	172	15	the	the	DET
cana-4260	172	16	classifier	classifier	NOUN
cana-4260	172	17	’s	’s	PART
cana-4260	172	18	ability	ability	NOUN
cana-4260	172	19	to	to	PART
cana-4260	172	20	generalize	generalize	VERB
cana-4260	172	21	by	by	ADP
cana-4260	172	22	producing	produce	VERB
cana-4260	172	23	better	well	ADJ
cana-4260	172	24	synthetic	synthetic	ADJ
cana-4260	172	25	samples	sample	NOUN
cana-4260	172	26	in	in	ADP
cana-4260	172	27	regions	region	NOUN
cana-4260	172	28	where	where	SCONJ
cana-4260	172	29	the	the	DET
cana-4260	172	30	minority	minority	NOUN
cana-4260	172	31	class	class	NOUN
cana-4260	172	32	is	be	AUX
cana-4260	172	33	underrepresented	underrepresented	ADJ
cana-4260	172	34	.	.	PUNCT
cana-4260	173	1	•	•	NUM
cana-4260	173	2	unlike	unlike	ADP
cana-4260	173	3	traditional	traditional	ADJ
cana-4260	173	4	smote	smote	NOUN
cana-4260	173	5	,	,	PUNCT
cana-4260	173	6	which	which	PRON
cana-4260	173	7	applies	apply	VERB
cana-4260	173	8	uniform	uniform	ADJ
cana-4260	173	9	sampling	sampling	NOUN
cana-4260	173	10	,	,	PUNCT
cana-4260	173	11	gmm	gmm	NOUN
cana-4260	173	12	-	-	PUNCT
cana-4260	173	13	smote	smote	ADJ
cana-4260	173	14	communications	communication	NOUN
cana-4260	173	15	on	on	ADP
cana-4260	173	16	applied	apply	VERB
cana-4260	173	17	nonlinear	nonlinear	ADJ
cana-4260	173	18	analysis	analysis	NOUN
cana-4260	173	19	issn	issn	NOUN
cana-4260	173	20	:	:	PUNCT
cana-4260	173	21	1074	1074	NUM
cana-4260	173	22	-	-	PUNCT
cana-4260	173	23	133x	133x	NUM
cana-4260	173	24	vol	vol	NOUN
cana-4260	173	25	32	32	NUM
cana-4260	173	26	no	no	NOUN
cana-4260	173	27	.	.	PUNCT
cana-4260	174	1	9s	9s	NUM
cana-4260	174	2	(	(	PUNCT
cana-4260	174	3	2025	2025	NUM
cana-4260	174	4	)	)	PUNCT
cana-4260	174	5	1639	1639	NUM
cana-4260	174	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-4260	174	7	adapts	adapt	VERB
cana-4260	174	8	to	to	ADP
cana-4260	174	9	the	the	DET
cana-4260	174	10	feature	feature	NOUN
cana-4260	174	11	distribution	distribution	NOUN
cana-4260	174	12	,	,	PUNCT
cana-4260	174	13	ensuring	ensure	VERB
cana-4260	174	14	that	that	SCONJ
cana-4260	174	15	oversampling	oversampling	NOUN
cana-4260	174	16	is	be	AUX
cana-4260	174	17	performed	perform	VERB
cana-4260	174	18	in	in	ADP
cana-4260	174	19	relevant	relevant	ADJ
cana-4260	174	20	data	datum	NOUN
cana-4260	174	21	regions	region	NOUN
cana-4260	174	22	.	.	PUNCT
cana-4260	175	1	•	•	NUM
cana-4260	175	2	the	the	DET
cana-4260	175	3	performance	performance	NOUN
cana-4260	175	4	improvement	improvement	NOUN
cana-4260	175	5	is	be	AUX
cana-4260	175	6	particularly	particularly	ADV
cana-4260	175	7	significant	significant	ADJ
cana-4260	175	8	in	in	ADP
cana-4260	175	9	highly	highly	ADV
cana-4260	175	10	imbalanced	imbalanced	ADJ
cana-4260	175	11	datasets	dataset	NOUN
cana-4260	175	12	,	,	PUNCT
cana-4260	175	13	where	where	SCONJ
cana-4260	175	14	simple	simple	ADJ
cana-4260	175	15	resampling	resampling	NOUN
cana-4260	175	16	methods	method	NOUN
cana-4260	175	17	tend	tend	VERB
cana-4260	175	18	to	to	PART
cana-4260	175	19	be	be	AUX
cana-4260	175	20	ineffective	ineffective	ADJ
cana-4260	175	21	.	.	PUNCT
cana-4260	176	1	potential	potential	ADJ
cana-4260	176	2	improvements	improvement	NOUN
cana-4260	176	3	:	:	PUNCT
cana-4260	176	4	•	•	NUM
cana-4260	176	5	adaptive	adaptive	ADJ
cana-4260	176	6	selection	selection	NOUN
cana-4260	176	7	of	of	ADP
cana-4260	176	8	the	the	DET
cana-4260	176	9	number	number	NOUN
cana-4260	176	10	of	of	ADP
cana-4260	176	11	clusters	cluster	NOUN
cana-4260	176	12	:	:	PUNCT
cana-4260	176	13	instead	instead	ADV
cana-4260	176	14	of	of	ADP
cana-4260	176	15	setting	set	VERB
cana-4260	176	16	a	a	DET
cana-4260	176	17	fixed	fix	VERB
cana-4260	176	18	number	number	NOUN
cana-4260	176	19	of	of	ADP
cana-4260	176	20	clusters	cluster	NOUN
cana-4260	176	21	(	(	PUNCT
cana-4260	176	22	k	k	X
cana-4260	176	23	in	in	ADP
cana-4260	176	24	gmm	gmm	PROPN
cana-4260	176	25	)	)	PUNCT
cana-4260	176	26	,	,	PUNCT
cana-4260	176	27	an	an	DET
cana-4260	176	28	automated	automate	VERB
cana-4260	176	29	selection	selection	NOUN
cana-4260	176	30	method	method	NOUN
cana-4260	176	31	(	(	PUNCT
cana-4260	176	32	such	such	ADJ
cana-4260	176	33	as	as	ADP
cana-4260	176	34	the	the	DET
cana-4260	176	35	bayesian	bayesian	NOUN
cana-4260	176	36	information	information	NOUN
cana-4260	176	37	criterion	criterion	NOUN
cana-4260	176	38	(	(	PUNCT
cana-4260	176	39	bic	bic	NOUN
cana-4260	176	40	)	)	PUNCT
cana-4260	176	41	or	or	CCONJ
cana-4260	176	42	the	the	DET
cana-4260	176	43	akaike	akaike	ADJ
cana-4260	176	44	information	information	NOUN
cana-4260	176	45	criterion	criterion	NOUN
cana-4260	176	46	(	(	PUNCT
cana-4260	176	47	aic	aic	PROPN
cana-4260	176	48	)	)	PUNCT
cana-4260	176	49	)	)	PUNCT
cana-4260	176	50	could	could	AUX
cana-4260	176	51	improve	improve	VERB
cana-4260	176	52	the	the	DET
cana-4260	176	53	clustering	clustering	ADJ
cana-4260	176	54	quality	quality	NOUN
cana-4260	176	55	.	.	PUNCT
cana-4260	177	1	•	•	NUM
cana-4260	177	2	hybrid	hybrid	ADJ
cana-4260	177	3	approach	approach	NOUN
cana-4260	177	4	with	with	ADP
cana-4260	177	5	deep	deep	ADJ
cana-4260	177	6	learning	learning	NOUN
cana-4260	177	7	:	:	PUNCT
cana-4260	177	8	combining	combine	VERB
cana-4260	177	9	gmm	gmm	NOUN
cana-4260	177	10	-	-	PUNCT
cana-4260	177	11	smote	smote	ADJ
cana-4260	177	12	with	with	ADP
cana-4260	177	13	ganbased	ganbase	VERB
cana-4260	177	14	(	(	PUNCT
cana-4260	177	15	generative	generative	ADJ
cana-4260	177	16	adversarial	adversarial	ADJ
cana-4260	177	17	networks	network	NOUN
cana-4260	177	18	)	)	PUNCT
cana-4260	177	19	synthetic	synthetic	ADJ
cana-4260	177	20	data	data	NOUN
cana-4260	177	21	generation	generation	NOUN
cana-4260	177	22	could	could	AUX
cana-4260	177	23	further	far	ADV
cana-4260	177	24	enhance	enhance	VERB
cana-4260	177	25	the	the	DET
cana-4260	177	26	quality	quality	NOUN
cana-4260	177	27	of	of	ADP
cana-4260	177	28	synthetic	synthetic	ADJ
cana-4260	177	29	samples	sample	NOUN
cana-4260	177	30	.	.	PUNCT
cana-4260	178	1	•	•	NUM
cana-4260	178	2	time	time	NOUN
cana-4260	178	3	complexity	complexity	NOUN
cana-4260	178	4	reduction	reduction	NOUN
cana-4260	178	5	:	:	PUNCT
cana-4260	178	6	gmm	gmm	PROPN
cana-4260	178	7	clustering	cluster	VERB
cana-4260	178	8	can	can	AUX
cana-4260	178	9	be	be	AUX
cana-4260	178	10	computationally	computationally	ADV
cana-4260	178	11	expensive	expensive	ADJ
cana-4260	178	12	for	for	ADP
cana-4260	178	13	large	large	ADJ
cana-4260	178	14	datasets	dataset	NOUN
cana-4260	178	15	.	.	PUNCT
cana-4260	179	1	optimizations	optimization	NOUN
cana-4260	179	2	such	such	ADJ
cana-4260	179	3	as	as	ADP
cana-4260	179	4	mini	mini	NOUN
cana-4260	179	5	-	-	NOUN
cana-4260	179	6	batch	batch	ADJ
cana-4260	179	7	gmm	gmm	NOUN
cana-4260	179	8	or	or	CCONJ
cana-4260	179	9	parallel	parallel	ADJ
cana-4260	179	10	computing	computing	NOUN
cana-4260	179	11	could	could	AUX
cana-4260	179	12	be	be	AUX
cana-4260	179	13	explored	explore	VERB
cana-4260	179	14	.	.	PUNCT
cana-4260	180	1	9	9	X
cana-4260	180	2	.	.	X
cana-4260	180	3	ablation	ablation	NOUN
cana-4260	180	4	study	study	NOUN
cana-4260	180	5	:	:	PUNCT
cana-4260	180	6	the	the	DET
cana-4260	180	7	ablation	ablation	NOUN
cana-4260	180	8	study	study	NOUN
cana-4260	180	9	examines	examine	VERB
cana-4260	180	10	the	the	DET
cana-4260	180	11	effect	effect	NOUN
cana-4260	180	12	of	of	ADP
cana-4260	180	13	various	various	ADJ
cana-4260	180	14	hyperparameters	hyperparameter	NOUN
cana-4260	180	15	on	on	ADP
cana-4260	180	16	performance	performance	NOUN
cana-4260	180	17	,	,	PUNCT
cana-4260	180	18	focusing	focus	VERB
cana-4260	180	19	on	on	ADP
cana-4260	180	20	how	how	SCONJ
cana-4260	180	21	changes	change	NOUN
cana-4260	180	22	impact	impact	NOUN
cana-4260	180	23	model	model	NOUN
cana-4260	180	24	effectiveness	effectiveness	NOUN
cana-4260	180	25	.	.	PUNCT
cana-4260	181	1	i.	i.	NOUN
cana-4260	181	2	number	number	NOUN
cana-4260	181	3	of	of	ADP
cana-4260	181	4	clusters	cluster	NOUN
cana-4260	181	5	(	(	PUNCT
cana-4260	181	6	k	k	X
cana-4260	181	7	in	in	ADP
cana-4260	181	8	gmm	gmm	PROPN
cana-4260	181	9	)	)	PUNCT
cana-4260	181	10	:	:	PUNCT
cana-4260	181	11	as	as	SCONJ
cana-4260	181	12	the	the	DET
cana-4260	181	13	number	number	NOUN
cana-4260	181	14	of	of	ADP
cana-4260	181	15	clusters	cluster	NOUN
cana-4260	181	16	increases	increase	NOUN
cana-4260	181	17	,	,	PUNCT
cana-4260	181	18	gmm	gmm	NOUN
cana-4260	181	19	-	-	PUNCT
cana-4260	181	20	smote	smote	ADJ
cana-4260	181	21	captures	capture	VERB
cana-4260	181	22	finer	fine	ADJ
cana-4260	181	23	variations	variation	NOUN
cana-4260	181	24	in	in	ADP
cana-4260	181	25	the	the	DET
cana-4260	181	26	data	datum	NOUN
cana-4260	181	27	.	.	PUNCT
cana-4260	182	1	however	however	ADV
cana-4260	182	2	,	,	PUNCT
cana-4260	182	3	too	too	ADV
cana-4260	182	4	many	many	ADJ
cana-4260	182	5	clusters	cluster	NOUN
cana-4260	182	6	may	may	AUX
cana-4260	182	7	lead	lead	VERB
cana-4260	182	8	to	to	ADP
cana-4260	182	9	over	over	ADP
cana-4260	182	10	-	-	PUNCT
cana-4260	182	11	segmentation	segmentation	NOUN
cana-4260	182	12	,	,	PUNCT
cana-4260	182	13	where	where	SCONJ
cana-4260	182	14	the	the	DET
cana-4260	182	15	minority	minority	NOUN
cana-4260	182	16	class	class	NOUN
cana-4260	182	17	is	be	AUX
cana-4260	182	18	split	split	VERB
cana-4260	182	19	too	too	ADV
cana-4260	182	20	finely	finely	ADV
cana-4260	182	21	,	,	PUNCT
cana-4260	182	22	reducing	reduce	VERB
cana-4260	182	23	the	the	DET
cana-4260	182	24	effectiveness	effectiveness	NOUN
cana-4260	182	25	of	of	ADP
cana-4260	182	26	synthetic	synthetic	ADJ
cana-4260	182	27	sample	sample	NOUN
cana-4260	182	28	generation	generation	NOUN
cana-4260	182	29	.	.	PUNCT
cana-4260	183	1	conversely	conversely	ADV
cana-4260	183	2	,	,	PUNCT
cana-4260	183	3	too	too	ADV
cana-4260	183	4	few	few	ADJ
cana-4260	183	5	clusters	cluster	NOUN
cana-4260	183	6	result	result	VERB
cana-4260	183	7	in	in	ADP
cana-4260	183	8	poor	poor	ADJ
cana-4260	183	9	data	datum	NOUN
cana-4260	183	10	representation	representation	NOUN
cana-4260	183	11	,	,	PUNCT
cana-4260	183	12	leading	lead	VERB
cana-4260	183	13	to	to	ADP
cana-4260	183	14	synthetic	synthetic	ADJ
cana-4260	183	15	samples	sample	NOUN
cana-4260	183	16	that	that	PRON
cana-4260	183	17	do	do	AUX
cana-4260	183	18	not	not	PART
cana-4260	183	19	generalize	generalize	VERB
cana-4260	183	20	well	well	ADV
cana-4260	183	21	.	.	PUNCT
cana-4260	184	1	ii	ii	PROPN
cana-4260	184	2	.	.	PUNCT
cana-4260	184	3	covariance	covariance	NOUN
cana-4260	184	4	type	type	NOUN
cana-4260	184	5	in	in	ADP
cana-4260	184	6	gmm	gmm	NOUN
cana-4260	184	7	:	:	PUNCT
cana-4260	184	8	full	full	ADJ
cana-4260	184	9	covariance	covariance	NOUN
cana-4260	184	10	matrices	matrix	NOUN
cana-4260	184	11	allow	allow	VERB
cana-4260	184	12	clusters	cluster	NOUN
cana-4260	184	13	to	to	PART
cana-4260	184	14	take	take	VERB
cana-4260	184	15	any	any	DET
cana-4260	184	16	shape	shape	NOUN
cana-4260	184	17	but	but	CCONJ
cana-4260	184	18	increase	increase	VERB
cana-4260	184	19	computational	computational	ADJ
cana-4260	184	20	cost	cost	NOUN
cana-4260	184	21	.	.	PUNCT
cana-4260	185	1	diagonal	diagonal	ADJ
cana-4260	185	2	covariance	covariance	NOUN
cana-4260	185	3	matrices	matrix	NOUN
cana-4260	185	4	speed	speed	VERB
cana-4260	185	5	up	up	ADP
cana-4260	185	6	training	training	NOUN
cana-4260	185	7	but	but	CCONJ
cana-4260	185	8	may	may	AUX
cana-4260	185	9	restrict	restrict	VERB
cana-4260	185	10	cluster	cluster	NOUN
cana-4260	185	11	flexibility	flexibility	NOUN
cana-4260	185	12	.	.	PUNCT
cana-4260	186	1	iii	iii	X
cana-4260	186	2	.	.	PUNCT
cana-4260	186	3	effect	effect	NOUN
cana-4260	186	4	of	of	ADP
cana-4260	186	5	oversampling	oversample	VERB
cana-4260	186	6	ratio	ratio	NOUN
cana-4260	186	7	:	:	PUNCT
cana-4260	186	8	increasing	increase	VERB
cana-4260	186	9	the	the	DET
cana-4260	186	10	oversampling	oversampling	ADJ
cana-4260	186	11	ratio	ratio	NOUN
cana-4260	186	12	beyond	beyond	ADP
cana-4260	186	13	a	a	DET
cana-4260	186	14	certain	certain	ADJ
cana-4260	186	15	point	point	NOUN
cana-4260	186	16	may	may	AUX
cana-4260	186	17	cause	cause	VERB
cana-4260	186	18	overfitting	overfitte	VERB
cana-4260	186	19	,	,	PUNCT
cana-4260	186	20	where	where	SCONJ
cana-4260	186	21	the	the	DET
cana-4260	186	22	model	model	NOUN
cana-4260	186	23	memorizes	memorize	VERB
cana-4260	186	24	synthetic	synthetic	ADJ
cana-4260	186	25	samples	sample	NOUN
cana-4260	186	26	rather	rather	ADV
cana-4260	186	27	than	than	ADP
cana-4260	186	28	generalizing	generalize	VERB
cana-4260	186	29	well	well	ADV
cana-4260	186	30	.	.	PUNCT
cana-4260	187	1	finding	find	VERB
cana-4260	187	2	the	the	DET
cana-4260	187	3	optimal	optimal	ADJ
cana-4260	187	4	oversampling	oversampling	ADJ
cana-4260	187	5	percentage	percentage	NOUN
cana-4260	187	6	is	be	AUX
cana-4260	187	7	crucial	crucial	ADJ
cana-4260	187	8	to	to	PART
cana-4260	187	9	balance	balance	VERB
cana-4260	187	10	minority	minority	NOUN
cana-4260	187	11	class	class	NOUN
cana-4260	187	12	representation	representation	NOUN
cana-4260	187	13	and	and	CCONJ
cana-4260	187	14	model	model	NOUN
cana-4260	187	15	generalization	generalization	NOUN
cana-4260	187	16	.	.	PUNCT
cana-4260	188	1	by	by	ADP
cana-4260	188	2	systematically	systematically	ADV
cana-4260	188	3	varying	vary	VERB
cana-4260	188	4	these	these	DET
cana-4260	188	5	hyperparameters	hyperparameter	NOUN
cana-4260	188	6	and	and	CCONJ
cana-4260	188	7	analyzing	analyze	VERB
cana-4260	188	8	their	their	PRON
cana-4260	188	9	impact	impact	NOUN
cana-4260	188	10	,	,	PUNCT
cana-4260	188	11	we	we	PRON
cana-4260	188	12	gain	gain	VERB
cana-4260	188	13	deeper	deep	ADJ
cana-4260	188	14	insights	insight	NOUN
cana-4260	188	15	into	into	ADP
cana-4260	188	16	the	the	DET
cana-4260	188	17	effectiveness	effectiveness	NOUN
cana-4260	188	18	of	of	ADP
cana-4260	188	19	gmm	gmm	NOUN
cana-4260	188	20	-	-	PUNCT
cana-4260	188	21	smote	smote	ADJ
cana-4260	188	22	and	and	CCONJ
cana-4260	188	23	identify	identify	VERB
cana-4260	188	24	the	the	DET
cana-4260	188	25	best	good	ADJ
cana-4260	188	26	configuration	configuration	NOUN
cana-4260	188	27	for	for	ADP
cana-4260	188	28	different	different	ADJ
cana-4260	188	29	types	type	NOUN
cana-4260	188	30	of	of	ADP
cana-4260	188	31	datasets	dataset	NOUN
cana-4260	188	32	.	.	PUNCT
cana-4260	189	1	10	10	X
cana-4260	189	2	.	.	X
cana-4260	189	3	conclusion	conclusion	NOUN
cana-4260	189	4	:	:	PUNCT
cana-4260	189	5	the	the	DET
cana-4260	189	6	study	study	NOUN
cana-4260	189	7	evaluates	evaluate	VERB
cana-4260	189	8	gmm	gmm	NOUN
cana-4260	189	9	-	-	PUNCT
cana-4260	189	10	smote	smote	ADJ
cana-4260	189	11	against	against	ADP
cana-4260	189	12	other	other	ADJ
cana-4260	189	13	oversampling	oversampling	ADJ
cana-4260	189	14	techniques	technique	NOUN
cana-4260	189	15	,	,	PUNCT
cana-4260	189	16	analyzing	analyze	VERB
cana-4260	189	17	its	its	PRON
cana-4260	189	18	impact	impact	NOUN
cana-4260	189	19	communications	communication	NOUN
cana-4260	189	20	on	on	ADP
cana-4260	189	21	applied	apply	VERB
cana-4260	189	22	nonlinear	nonlinear	ADJ
cana-4260	189	23	analysis	analysis	NOUN
cana-4260	189	24	issn	issn	NOUN
cana-4260	189	25	:	:	PUNCT
cana-4260	189	26	1074	1074	NUM
cana-4260	189	27	-	-	PUNCT
cana-4260	189	28	133x	133x	NUM
cana-4260	189	29	vol	vol	NOUN
cana-4260	189	30	32	32	NUM
cana-4260	189	31	no	no	NOUN
cana-4260	189	32	.	.	PUNCT
cana-4260	190	1	9s	9s	NUM
cana-4260	190	2	(	(	PUNCT
cana-4260	190	3	2025	2025	NUM
cana-4260	190	4	)	)	PUNCT
cana-4260	190	5	1640	1640	NUM
cana-4260	190	6	https://internationalpubls.com	https://internationalpubls.com	X
cana-4260	190	7	on	on	ADP
cana-4260	190	8	multiple	multiple	ADJ
cana-4260	190	9	datasets	dataset	NOUN
cana-4260	190	10	and	and	CCONJ
cana-4260	190	11	classifiers	classifier	NOUN
cana-4260	190	12	.	.	PUNCT
cana-4260	191	1	performance	performance	NOUN
cana-4260	191	2	metrics	metric	NOUN
cana-4260	191	3	such	such	ADJ
cana-4260	191	4	as	as	ADP
cana-4260	191	5	accuracy	accuracy	NOUN
cana-4260	191	6	,	,	PUNCT
cana-4260	191	7	auc	auc	NOUN
cana-4260	191	8	-	-	PUNCT
cana-4260	191	9	roc	roc	NOUN
cana-4260	191	10	,	,	PUNCT
cana-4260	191	11	and	and	CCONJ
cana-4260	191	12	k	k	PROPN
cana-4260	191	13	value	value	NOUN
cana-4260	191	14	demonstrate	demonstrate	VERB
cana-4260	191	15	that	that	SCONJ
cana-4260	191	16	gmm	gmm	NOUN
cana-4260	191	17	-	-	PUNCT
cana-4260	191	18	smote	smote	ADJ
cana-4260	191	19	effectively	effectively	ADV
cana-4260	191	20	enhances	enhance	VERB
cana-4260	191	21	minority	minority	NOUN
cana-4260	191	22	class	class	NOUN
cana-4260	191	23	representation	representation	NOUN
cana-4260	191	24	while	while	SCONJ
cana-4260	191	25	maintaining	maintain	VERB
cana-4260	191	26	a	a	DET
cana-4260	191	27	balanced	balanced	ADJ
cana-4260	191	28	decision	decision	NOUN
cana-4260	191	29	boundary	boundary	NOUN
cana-4260	191	30	.	.	PUNCT
cana-4260	192	1	the	the	DET
cana-4260	192	2	results	result	NOUN
cana-4260	192	3	confirm	confirm	VERB
cana-4260	192	4	its	its	PRON
cana-4260	192	5	superiority	superiority	NOUN
cana-4260	192	6	over	over	ADP
cana-4260	192	7	traditional	traditional	ADJ
cana-4260	192	8	smote	smote	NOUN
cana-4260	192	9	and	and	CCONJ
cana-4260	192	10	adasyn	adasyn	NOUN
cana-4260	192	11	in	in	ADP
cana-4260	192	12	generating	generate	VERB
cana-4260	192	13	meaningful	meaningful	ADJ
cana-4260	192	14	synthetic	synthetic	ADJ
cana-4260	192	15	samples	sample	NOUN
cana-4260	192	16	and	and	CCONJ
cana-4260	192	17	reducing	reduce	VERB
cana-4260	192	18	overfitting	overfitting	NOUN
cana-4260	192	19	.	.	PUNCT
cana-4260	193	1	future	future	ADJ
cana-4260	193	2	work	work	NOUN
cana-4260	193	3	can	can	AUX
cana-4260	193	4	explore	explore	VERB
cana-4260	193	5	adaptive	adaptive	ADJ
cana-4260	193	6	parameter	parameter	NOUN
cana-4260	193	7	tuning	tune	VERB
cana-4260	193	8	for	for	ADP
cana-4260	193	9	gmmsmote	gmmsmote	NOUN
cana-4260	193	10	to	to	PART
cana-4260	193	11	optimize	optimize	VERB
cana-4260	193	12	performance	performance	NOUN
cana-4260	193	13	across	across	ADP
cana-4260	193	14	diverse	diverse	ADJ
cana-4260	193	15	datasets	dataset	NOUN
cana-4260	193	16	.	.	PUNCT
cana-4260	194	1	additionally	additionally	ADV
cana-4260	194	2	,	,	PUNCT
cana-4260	194	3	integrating	integrate	VERB
cana-4260	194	4	gmm	gmm	NOUN
cana-4260	194	5	-	-	PUNCT
cana-4260	194	6	smote	smote	ADJ
cana-4260	194	7	with	with	ADP
cana-4260	194	8	deep	deep	ADJ
cana-4260	194	9	learning	learning	NOUN
cana-4260	194	10	models	model	NOUN
cana-4260	194	11	and	and	CCONJ
cana-4260	194	12	extending	extend	VERB
cana-4260	194	13	its	its	PRON
cana-4260	194	14	application	application	NOUN
cana-4260	194	15	to	to	ADP
cana-4260	194	16	real	real	ADJ
cana-4260	194	17	-	-	PUNCT
cana-4260	194	18	time	time	NOUN
cana-4260	194	19	imbalanced	imbalanced	ADJ
cana-4260	194	20	data	data	NOUN
cana-4260	194	21	scenarios	scenario	NOUN
cana-4260	194	22	could	could	AUX
cana-4260	194	23	further	far	ADV
cana-4260	194	24	enhance	enhance	VERB
cana-4260	194	25	its	its	PRON
cana-4260	194	26	effectiveness	effectiveness	NOUN
cana-4260	194	27	.	.	PUNCT
cana-4260	195	1	evaluating	evaluate	VERB
cana-4260	195	2	its	its	PRON
cana-4260	195	3	impact	impact	NOUN
cana-4260	195	4	on	on	ADP
cana-4260	195	5	multi	multi	ADJ
cana-4260	195	6	-	-	ADJ
cana-4260	195	7	class	class	ADJ
cana-4260	195	8	imbalanced	imbalanced	ADJ
cana-4260	195	9	problems	problem	NOUN
cana-4260	195	10	and	and	CCONJ
cana-4260	195	11	refining	refine	VERB
cana-4260	195	12	computational	computational	ADJ
cana-4260	195	13	efficiency	efficiency	NOUN
cana-4260	195	14	are	be	AUX
cana-4260	195	15	also	also	ADV
cana-4260	195	16	promising	promise	VERB
cana-4260	195	17	directions	direction	NOUN
cana-4260	195	18	.	.	PUNCT
cana-4260	196	1	this	this	DET
cana-4260	196	2	paper	paper	NOUN
cana-4260	196	3	presents	presents	AUX
cana-4260	196	4	gmm	gmm	NOUN
cana-4260	196	5	-	-	PUNCT
cana-4260	196	6	smote	smote	ADJ
cana-4260	196	7	as	as	ADP
cana-4260	196	8	a	a	DET
cana-4260	196	9	robust	robust	ADJ
cana-4260	196	10	resampling	resampling	NOUN
cana-4260	196	11	technique	technique	NOUN
cana-4260	196	12	that	that	PRON
cana-4260	196	13	combines	combine	VERB
cana-4260	196	14	the	the	DET
cana-4260	196	15	benefits	benefit	NOUN
cana-4260	196	16	of	of	ADP
cana-4260	196	17	gaussian	gaussian	ADJ
cana-4260	196	18	mixture	mixture	NOUN
cana-4260	196	19	models	model	NOUN
cana-4260	196	20	and	and	CCONJ
cana-4260	196	21	smote	smote	VERB
cana-4260	196	22	to	to	PART
cana-4260	196	23	improve	improve	VERB
cana-4260	196	24	classification	classification	NOUN
cana-4260	196	25	performance	performance	NOUN
cana-4260	196	26	on	on	ADP
cana-4260	196	27	imbalanced	imbalanced	ADJ
cana-4260	196	28	datasets	dataset	NOUN
cana-4260	196	29	.	.	PUNCT
cana-4260	197	1	through	through	ADP
cana-4260	197	2	extensive	extensive	ADJ
cana-4260	197	3	experiments	experiment	NOUN
cana-4260	197	4	,	,	PUNCT
cana-4260	197	5	we	we	PRON
cana-4260	197	6	demonstrate	demonstrate	VERB
cana-4260	197	7	its	its	PRON
cana-4260	197	8	effectiveness	effectiveness	NOUN
cana-4260	197	9	,	,	PUNCT
cana-4260	197	10	highlighting	highlight	VERB
cana-4260	197	11	its	its	PRON
cana-4260	197	12	potential	potential	NOUN
cana-4260	197	13	for	for	ADP
cana-4260	197	14	future	future	ADJ
cana-4260	197	15	advancements	advancement	NOUN
cana-4260	197	16	in	in	ADP
cana-4260	197	17	machine	machine	NOUN
cana-4260	197	18	learning	learning	NOUN
cana-4260	197	19	and	and	CCONJ
cana-4260	197	20	imbalanced	imbalanced	ADJ
cana-4260	197	21	data	datum	NOUN
cana-4260	197	22	handling	handling	NOUN
cana-4260	197	23	.	.	PUNCT
cana-4260	198	1	refrences	refrence	VERB
cana-4260	199	1	[	[	X
cana-4260	199	2	1	1	X
cana-4260	199	3	]	]	X
cana-4260	199	4	lorem	lorem	PROPN
cana-4260	199	5	ipsum	ipsum	PROPN
cana-4260	199	6	dolor	dolor	PROPN
cana-4260	199	7	sit	sit	PROPN
cana-4260	199	8	amet	amet	PROPN
cana-4260	199	9	,	,	PUNCT
cana-4260	199	10	consectetur	consectetur	PROPN
cana-4260	199	11	adipiscing	adipiscing	PROPN
cana-4260	199	12	elit	elit	PROPN
cana-4260	199	13	,	,	PUNCT
cana-4260	199	14	sed	se	VERB
cana-4260	199	15	do	do	VERB
cana-4260	199	16	eiusmod	eiusmod	PROPN
cana-4260	199	17	tempor	tempor	PROPN
cana-4260	199	18	incididunt	incididunt	PROPN
cana-4260	199	19	ut	ut	PROPN
cana-4260	199	20	labore	labore	PROPN
cana-4260	199	21	et	et	PROPN
cana-4260	199	22	dolore	dolore	PROPN
cana-4260	199	23	magna	magna	PROPN
cana-4260	199	24	aliqua	aliqua	PROPN
cana-4260	199	25	.	.	PUNCT
cana-4260	200	1	[	[	X
cana-4260	200	2	2	2	NUM
cana-4260	200	3	]	]	PUNCT
cana-4260	200	4	ipsum	ipsum	PROPN
cana-4260	200	5	dolor	dolor	PROPN
cana-4260	200	6	sit	sit	PROPN
cana-4260	200	7	amet	amet	PROPN
cana-4260	200	8	consectetur	consectetur	PROPN
cana-4260	200	9	adipiscing	adipiscing	PROPN
cana-4260	200	10	elit	elit	PROPN
cana-4260	200	11	pellentesque	pellentesque	NOUN
cana-4260	200	12	.	.	PUNCT
cana-4260	201	1	orci	orci	PROPN
cana-4260	201	2	eu	eu	PROPN
cana-4260	201	3	lobortis	lobortis	PROPN
cana-4260	201	4	elementum	elementum	PROPN
cana-4260	201	5	nibh	nibh	PROPN
cana-4260	201	6	.	.	PUNCT
cana-4260	202	1	faucibus	faucibus	VERB
cana-4260	202	2	a	a	DET
cana-4260	202	3	pellentesque	pellentesque	NOUN
cana-4260	202	4	sit	sit	NOUN
cana-4260	202	5	amet	amet	NOUN
cana-4260	202	6	porttitor	porttitor	NOUN
cana-4260	202	7	.	.	PUNCT
cana-4260	203	1	[	[	X
cana-4260	203	2	3	3	NUM
cana-4260	203	3	]	]	X
cana-4260	203	4	egestas	egesta	NOUN
cana-4260	203	5	tellus	tellus	ADJ
cana-4260	203	6	rutrum	rutrum	X
cana-4260	203	7	tellus	tellus	ADJ
cana-4260	203	8	pellentesque	pellentesque	PROPN
cana-4260	203	9	eu	eu	PROPN
cana-4260	203	10	tincidunt	tincidunt	PROPN
cana-4260	203	11	tortor	tortor	NOUN
cana-4260	203	12	.	.	PUNCT
cana-4260	204	1	sagittis	sagittis	PROPN
cana-4260	204	2	orci	orci	PROPN
cana-4260	204	3	a	a	DET
cana-4260	204	4	scelerisque	scelerisque	ADJ
cana-4260	204	5	purus	purus	PROPN
cana-4260	204	6	semper	semper	PROPN
cana-4260	204	7	eget	eget	PROPN
cana-4260	204	8	.	.	PUNCT
cana-4260	205	1	vitae	vitae	PROPN
cana-4260	205	2	purus	purus	PROPN
cana-4260	205	3	faucibus	faucibus	PROPN
cana-4260	205	4	ornare	ornare	PROPN
cana-4260	205	5	suspendisse	suspendisse	PROPN
cana-4260	205	6	sed	sed	PROPN
cana-4260	205	7	nisi	nisi	PROPN
cana-4260	205	8	lacus	lacus	PROPN
cana-4260	205	9	sed	sed	PROPN
cana-4260	205	10	viverra	viverra	PROPN
cana-4260	205	11	.	.	PUNCT
cana-4260	206	1	[	[	X
cana-4260	206	2	4	4	NUM
cana-4260	206	3	]	]	PUNCT
cana-4260	206	4	augue	augue	NOUN
cana-4260	206	5	interdum	interdum	PROPN
cana-4260	206	6	velit	velit	PROPN
cana-4260	206	7	euismod	euismod	PROPN
cana-4260	206	8	in	in	ADP
cana-4260	206	9	pellentesque	pellentesque	ADJ
cana-4260	206	10	massa	massa	PROPN
cana-4260	206	11	placerat	placerat	PROPN
cana-4260	206	12	duis	duis	PROPN
cana-4260	206	13	ultricies	ultricies	PROPN
cana-4260	206	14	.	.	PUNCT
cana-4260	207	1	metus	metus	PROPN
cana-4260	207	2	aliquam	aliquam	PROPN
cana-4260	207	3	eleifend	eleifend	PROPN
cana-4260	207	4	mi	mi	PROPN
cana-4260	207	5	in	in	ADP
cana-4260	207	6	nulla	nulla	PROPN
cana-4260	207	7	posuere	posuere	ADV
cana-4260	207	8	sollicitudin	sollicitudin	VERB
cana-4260	207	9	aliquam	aliquam	NOUN
cana-4260	207	10	ultrices	ultrice	NOUN
cana-4260	207	11	.	.	PUNCT
cana-4260	208	1	[	[	X
cana-4260	208	2	5	5	NUM
cana-4260	208	3	]	]	PUNCT
cana-4260	208	4	velit	velit	PROPN
cana-4260	208	5	laoreet	laoreet	PROPN
cana-4260	208	6	i	i	PROPN
cana-4260	208	7	d	d	PROPN
cana-4260	208	8	donec	donec	PROPN
cana-4260	208	9	ultrices	ultrice	NOUN
cana-4260	208	10	tincidunt	tincidunt	PROPN
cana-4260	208	11	arcu	arcu	PROPN
cana-4260	208	12	non	non	PROPN
cana-4260	208	13	sodales	sodales	PROPN
cana-4260	208	14	neque	neque	PROPN
cana-4260	208	15	.	.	PUNCT
cana-4260	209	1	non	non	PROPN
cana-4260	209	2	curabitur	curabitur	PROPN
cana-4260	209	3	gravida	gravida	PROPN
cana-4260	209	4	arcu	arcu	PROPN
cana-4260	209	5	ac	ac	PROPN
cana-4260	209	6	tortor	tortor	PROPN
cana-4260	209	7	dignissim	dignissim	PROPN
cana-4260	209	8	convallis	convalli	NOUN
cana-4260	209	9	aenean	aenean	PROPN
cana-4260	209	10	et	et	NOUN
cana-4260	209	11	.	.	PUNCT
cana-4260	210	1	[	[	X
cana-4260	210	2	6	6	NUM
cana-4260	210	3	]	]	PUNCT
cana-4260	210	4	euismod	euismod	NOUN
cana-4260	210	5	in	in	ADP
cana-4260	210	6	pellentesque	pellentesque	ADJ
cana-4260	210	7	massa	massa	PROPN
cana-4260	210	8	placerat	placerat	NOUN
cana-4260	210	9	.	.	PUNCT
cana-4260	211	1	morbi	morbi	NOUN
cana-4260	211	2	non	non	PROPN
cana-4260	212	1	arcu	arcu	PROPN
cana-4260	212	2	risus	risus	PROPN
cana-4260	212	3	quis	quis	PROPN
cana-4260	212	4	varius	varius	PROPN
cana-4260	212	5	quam	quam	PROPN
cana-4260	212	6	quisque	quisque	PROPN
cana-4260	212	7	.	.	PUNCT