id	sid	tid	token	lemma	pos
bracis-33630	1	1	evaluating	evaluate	VERB
bracis-33630	1	2	short	short	ADJ
bracis-33630	1	3	text	text	NOUN
bracis-33630	1	4	stream	stream	NOUN
bracis-33630	1	5	clustering	cluster	VERB
bracis-33630	1	6	on	on	ADP
bracis-33630	1	7	 	 	SPACE
bracis-33630	1	8	large	large	ADJ
bracis-33630	1	9	e	e	ADJ
bracis-33630	1	10	-	-	NOUN
bracis-33630	1	11	commerce	commerce	NOUN
bracis-33630	1	12	datasets	dataset	NOUN
bracis-33630	1	13	|	|	ADV
bracis-33630	1	14	springer	springer	NOUN
bracis-33630	1	15	nature	nature	PROPN
bracis-33630	1	16	link	link	PROPN
bracis-33630	1	17	(	(	PUNCT
bracis-33630	1	18	formerly	formerly	ADV
bracis-33630	1	19	springerlink	springerlink	NOUN
bracis-33630	1	20	)	)	PUNCT
bracis-33630	1	21	skip	skip	VERB
bracis-33630	1	22	to	to	ADP
bracis-33630	1	23	main	main	ADJ
bracis-33630	1	24	content	content	NOUN
bracis-33630	1	25	advertisement	advertisement	NOUN
bracis-33630	1	26	log	log	NOUN
bracis-33630	1	27	in	in	ADP
bracis-33630	1	28	menu	menu	NOUN
bracis-33630	1	29	find	find	VERB
bracis-33630	1	30	a	a	DET
bracis-33630	1	31	journal	journal	NOUN
bracis-33630	1	32	publish	publish	VERB
bracis-33630	1	33	with	with	ADP
bracis-33630	1	34	us	we	PRON
bracis-33630	1	35	track	track	VERB
bracis-33630	1	36	your	your	PRON
bracis-33630	1	37	research	research	NOUN
bracis-33630	1	38	search	search	NOUN
bracis-33630	1	39	cart	cart	NOUN
bracis-33630	1	40	home	home	NOUN
bracis-33630	1	41	intelligent	intelligent	ADJ
bracis-33630	1	42	systems	system	NOUN
bracis-33630	1	43	conference	conference	NOUN
bracis-33630	1	44	paper	paper	NOUN
bracis-33630	1	45	evaluating	evaluate	VERB
bracis-33630	1	46	short	short	ADJ
bracis-33630	1	47	text	text	NOUN
bracis-33630	1	48	stream	stream	NOUN
bracis-33630	1	49	clustering	cluster	VERB
bracis-33630	1	50	on	on	ADP
bracis-33630	1	51	 	 	SPACE
bracis-33630	1	52	large	large	ADJ
bracis-33630	1	53	e	e	ADJ
bracis-33630	1	54	-	-	NOUN
bracis-33630	1	55	commerce	commerce	NOUN
bracis-33630	1	56	datasets	dataset	NOUN
bracis-33630	1	57	conference	conference	NOUN
bracis-33630	1	58	paper	paper	NOUN
bracis-33630	1	59	first	first	ADV
bracis-33630	1	60	online	online	ADV
bracis-33630	1	61	:	:	PUNCT
bracis-33630	1	62	30	30	NUM
bracis-33630	1	63	january	january	NOUN
bracis-33630	1	64	2025	2025	NUM
bracis-33630	1	65	pp	pp	PROPN
bracis-33630	1	66	245–259	245–259	NUM
bracis-33630	1	67	cite	cite	VERB
bracis-33630	1	68	this	this	DET
bracis-33630	1	69	conference	conference	NOUN
bracis-33630	1	70	paper	paper	NOUN
bracis-33630	1	71	access	access	NOUN
bracis-33630	1	72	provided	provide	VERB
bracis-33630	1	73	by	by	ADP
bracis-33630	1	74	university	university	PROPN
bracis-33630	1	75	of	of	ADP
bracis-33630	1	76	notre	notre	PROPN
bracis-33630	1	77	dame	dame	PROPN
bracis-33630	1	78	hesburgh	hesburgh	PROPN
bracis-33630	1	79	library	library	PROPN
bracis-33630	1	80	download	download	PROPN
bracis-33630	1	81	book	book	NOUN
bracis-33630	1	82	pdf	pdf	PROPN
bracis-33630	1	83	download	download	NOUN
bracis-33630	1	84	book	book	NOUN
bracis-33630	1	85	epub	epub	PROPN
bracis-33630	1	86	intelligent	intelligent	ADJ
bracis-33630	1	87	systems	system	NOUN
bracis-33630	1	88	(	(	PUNCT
bracis-33630	1	89	bracis	bracis	NOUN
bracis-33630	1	90	2024	2024	NUM
bracis-33630	1	91	)	)	PUNCT
bracis-33630	1	92	evaluating	evaluate	VERB
bracis-33630	1	93	short	short	ADJ
bracis-33630	1	94	text	text	NOUN
bracis-33630	1	95	stream	stream	NOUN
bracis-33630	1	96	clustering	cluster	VERB
bracis-33630	1	97	on	on	ADP
bracis-33630	1	98	 	 	SPACE
bracis-33630	1	99	large	large	ADJ
bracis-33630	1	100	e	e	ADJ
bracis-33630	1	101	-	-	NOUN
bracis-33630	1	102	commerce	commerce	NOUN
bracis-33630	1	103	datasets	dataset	NOUN
bracis-33630	1	104	download	download	NOUN
bracis-33630	1	105	book	book	NOUN
bracis-33630	1	106	pdf	pdf	PROPN
bracis-33630	1	107	download	download	NOUN
bracis-33630	1	108	book	book	PROPN
bracis-33630	1	109	epub	epub	PROPN
bracis-33630	1	110	cesar	cesar	PROPN
bracis-33630	1	111	andrade	andrade	PROPN
bracis-33630	1	112	  	  	SPACE
bracis-33630	1	113	orcid	orcid	PROPN
bracis-33630	1	114	:	:	PUNCT
bracis-33630	1	115	orcid.org/0009-0005-2218-66489,12	orcid.org/0009-0005-2218-66489,12	PROPN
bracis-33630	1	116	,	,	PUNCT
bracis-33630	2	1	rita	rita	PROPN
bracis-33630	2	2	p.	p.	PROPN
bracis-33630	2	3	ribeiro	ribeiro	PROPN
bracis-33630	2	4	  	  	SPACE
bracis-33630	2	5	orcid	orcid	PROPN
bracis-33630	2	6	:	:	PUNCT
bracis-33630	2	7	orcid.org/0000-0002-6852-80779,12	orcid.org/0000-0002-6852-80779,12	PROPN
bracis-33630	2	8	&	&	CCONJ
bracis-33630	2	9	joão	joão	PROPN
bracis-33630	2	10	gama	gama	PROPN
bracis-33630	2	11	  	  	SPACE
bracis-33630	2	12	orcid	orcid	NOUN
bracis-33630	2	13	:	:	PUNCT
bracis-33630	2	14	orcid.org/0000-0003-3357-119510,11,12	orcid.org/0000-0003-3357-119510,11,12	NOUN
bracis-33630	2	15	  	  	SPACE
bracis-33630	2	16	part	part	NOUN
bracis-33630	2	17	of	of	ADP
bracis-33630	2	18	the	the	DET
bracis-33630	2	19	book	book	NOUN
bracis-33630	2	20	series	series	NOUN
bracis-33630	2	21	:	:	PUNCT
bracis-33630	2	22	lecture	lecture	NOUN
bracis-33630	2	23	notes	note	NOUN
bracis-33630	2	24	in	in	ADP
bracis-33630	2	25	computer	computer	NOUN
bracis-33630	2	26	science	science	NOUN
bracis-33630	2	27	(	(	PUNCT
bracis-33630	2	28	(	(	PUNCT
bracis-33630	2	29	lnai	lnai	ADJ
bracis-33630	2	30	,	,	PUNCT
bracis-33630	2	31	volume	volume	NOUN
bracis-33630	2	32	15414	15414	NUM
bracis-33630	2	33	)	)	PUNCT
bracis-33630	2	34	)	)	PUNCT
bracis-33630	2	35	included	include	VERB
bracis-33630	2	36	in	in	ADP
bracis-33630	2	37	the	the	DET
bracis-33630	2	38	following	follow	VERB
bracis-33630	2	39	conference	conference	NOUN
bracis-33630	2	40	series	series	NOUN
bracis-33630	2	41	:	:	PUNCT
bracis-33630	2	42	brazilian	brazilian	ADJ
bracis-33630	2	43	conference	conference	NOUN
bracis-33630	2	44	on	on	ADP
bracis-33630	2	45	intelligent	intelligent	ADJ
bracis-33630	2	46	systems	system	NOUN
bracis-33630	2	47	372	372	NUM
bracis-33630	2	48	accesses	access	NOUN
bracis-33630	2	49	1	1	NUM
bracis-33630	2	50	citation	citation	NOUN
bracis-33630	2	51	abstract	abstract	ADJ
bracis-33630	2	52	latent	latent	PROPN
bracis-33630	2	53	dirichlet	dirichlet	PROPN
bracis-33630	2	54	allocation	allocation	NOUN
bracis-33630	2	55	(	(	PUNCT
bracis-33630	2	56	lda	lda	PROPN
bracis-33630	2	57	)	)	PUNCT
bracis-33630	2	58	is	be	AUX
bracis-33630	2	59	a	a	DET
bracis-33630	2	60	fundamental	fundamental	ADJ
bracis-33630	2	61	method	method	NOUN
bracis-33630	2	62	for	for	ADP
bracis-33630	2	63	clustering	cluster	VERB
bracis-33630	2	64	short	short	ADJ
bracis-33630	2	65	text	text	NOUN
bracis-33630	2	66	streams	stream	NOUN
bracis-33630	2	67	.	.	PUNCT
bracis-33630	3	1	however	however	ADV
bracis-33630	3	2	,	,	PUNCT
bracis-33630	3	3	when	when	SCONJ
bracis-33630	3	4	applied	apply	VERB
bracis-33630	3	5	to	to	ADP
bracis-33630	3	6	large	large	ADJ
bracis-33630	3	7	datasets	dataset	NOUN
bracis-33630	3	8	,	,	PUNCT
bracis-33630	3	9	it	it	PRON
bracis-33630	3	10	often	often	ADV
bracis-33630	3	11	faces	face	VERB
bracis-33630	3	12	significant	significant	ADJ
bracis-33630	3	13	challenges	challenge	NOUN
bracis-33630	3	14	,	,	PUNCT
bracis-33630	3	15	and	and	CCONJ
bracis-33630	3	16	its	its	PRON
bracis-33630	3	17	performance	performance	NOUN
bracis-33630	3	18	is	be	AUX
bracis-33630	3	19	typically	typically	ADV
bracis-33630	3	20	evaluated	evaluate	VERB
bracis-33630	3	21	in	in	ADP
bracis-33630	3	22	domain	domain	NOUN
bracis-33630	3	23	-	-	PUNCT
bracis-33630	3	24	specific	specific	ADJ
bracis-33630	3	25	datasets	dataset	NOUN
bracis-33630	3	26	such	such	ADJ
bracis-33630	3	27	as	as	ADP
bracis-33630	3	28	news	news	NOUN
bracis-33630	3	29	and	and	CCONJ
bracis-33630	3	30	tweets	tweet	NOUN
bracis-33630	3	31	.	.	PUNCT
bracis-33630	4	1	this	this	DET
bracis-33630	4	2	study	study	NOUN
bracis-33630	4	3	aims	aim	VERB
bracis-33630	4	4	to	to	PART
bracis-33630	4	5	fill	fill	VERB
bracis-33630	4	6	this	this	DET
bracis-33630	4	7	gap	gap	NOUN
bracis-33630	4	8	by	by	ADP
bracis-33630	4	9	evaluating	evaluate	VERB
bracis-33630	4	10	the	the	DET
bracis-33630	4	11	effectiveness	effectiveness	NOUN
bracis-33630	4	12	of	of	ADP
bracis-33630	4	13	short	short	ADJ
bracis-33630	4	14	text	text	NOUN
bracis-33630	4	15	clustering	cluster	VERB
bracis-33630	4	16	methods	method	NOUN
bracis-33630	4	17	in	in	ADP
bracis-33630	4	18	a	a	DET
bracis-33630	4	19	large	large	ADJ
bracis-33630	4	20	and	and	CCONJ
bracis-33630	4	21	diverse	diverse	ADJ
bracis-33630	4	22	e	e	NOUN
bracis-33630	4	23	-	-	NOUN
bracis-33630	4	24	commerce	commerce	NOUN
bracis-33630	4	25	dataset	dataset	NOUN
bracis-33630	4	26	.	.	PUNCT
bracis-33630	5	1	we	we	PRON
bracis-33630	5	2	specifically	specifically	ADV
bracis-33630	5	3	investigate	investigate	VERB
bracis-33630	5	4	how	how	SCONJ
bracis-33630	5	5	well	well	ADV
bracis-33630	5	6	these	these	DET
bracis-33630	5	7	clustering	cluster	VERB
bracis-33630	5	8	algorithms	algorithm	NOUN
bracis-33630	5	9	adapt	adapt	VERB
bracis-33630	5	10	to	to	ADP
bracis-33630	5	11	the	the	DET
bracis-33630	5	12	complex	complex	ADJ
bracis-33630	5	13	dynamics	dynamic	NOUN
bracis-33630	5	14	and	and	CCONJ
bracis-33630	5	15	larger	large	ADJ
bracis-33630	5	16	scale	scale	NOUN
bracis-33630	5	17	of	of	ADP
bracis-33630	5	18	e	e	NOUN
bracis-33630	5	19	-	-	NOUN
bracis-33630	5	20	commerce	commerce	NOUN
bracis-33630	5	21	text	text	NOUN
bracis-33630	5	22	streams	stream	NOUN
bracis-33630	5	23	,	,	PUNCT
bracis-33630	5	24	which	which	PRON
bracis-33630	5	25	differ	differ	VERB
bracis-33630	5	26	from	from	ADP
bracis-33630	5	27	their	their	PRON
bracis-33630	5	28	usual	usual	ADJ
bracis-33630	5	29	application	application	NOUN
bracis-33630	5	30	domains	domain	NOUN
bracis-33630	5	31	.	.	PUNCT
bracis-33630	6	1	our	our	PRON
bracis-33630	6	2	analysis	analysis	NOUN
bracis-33630	6	3	focuses	focus	VERB
bracis-33630	6	4	on	on	ADP
bracis-33630	6	5	the	the	DET
bracis-33630	6	6	impact	impact	NOUN
bracis-33630	6	7	of	of	ADP
bracis-33630	6	8	high	high	ADJ
bracis-33630	6	9	homogeneity	homogeneity	NOUN
bracis-33630	6	10	scores	score	NOUN
bracis-33630	6	11	on	on	ADP
bracis-33630	6	12	the	the	DET
bracis-33630	6	13	reported	report	VERB
bracis-33630	6	14	normalized	normalize	VERB
bracis-33630	6	15	mutual	mutual	ADJ
bracis-33630	6	16	information	information	NOUN
bracis-33630	6	17	(	(	PUNCT
bracis-33630	6	18	nmi	nmi	NOUN
bracis-33630	6	19	)	)	PUNCT
bracis-33630	6	20	values	value	NOUN
bracis-33630	6	21	.	.	PUNCT
bracis-33630	7	1	we	we	PRON
bracis-33630	7	2	particularly	particularly	ADV
bracis-33630	7	3	examine	examine	VERB
bracis-33630	7	4	whether	whether	SCONJ
bracis-33630	7	5	these	these	DET
bracis-33630	7	6	scores	score	NOUN
bracis-33630	7	7	are	be	AUX
bracis-33630	7	8	inflated	inflate	VERB
bracis-33630	7	9	due	due	ADP
bracis-33630	7	10	to	to	ADP
bracis-33630	7	11	the	the	DET
bracis-33630	7	12	prevalence	prevalence	NOUN
bracis-33630	7	13	of	of	ADP
bracis-33630	7	14	single	single	ADJ
bracis-33630	7	15	-	-	PUNCT
bracis-33630	7	16	element	element	NOUN
bracis-33630	7	17	clusters	cluster	NOUN
bracis-33630	7	18	.	.	PUNCT
bracis-33630	8	1	to	to	PART
bracis-33630	8	2	address	address	VERB
bracis-33630	8	3	potential	potential	ADJ
bracis-33630	8	4	biases	bias	NOUN
bracis-33630	8	5	in	in	ADP
bracis-33630	8	6	clustering	clustering	ADJ
bracis-33630	8	7	evaluation	evaluation	NOUN
bracis-33630	8	8	,	,	PUNCT
bracis-33630	8	9	we	we	PRON
bracis-33630	8	10	propose	propose	VERB
bracis-33630	8	11	using	use	VERB
bracis-33630	8	12	the	the	DET
bracis-33630	8	13	akaike	akaike	ADJ
bracis-33630	8	14	information	information	NOUN
bracis-33630	8	15	criterion	criterion	NOUN
bracis-33630	8	16	(	(	PUNCT
bracis-33630	8	17	aic	aic	PROPN
bracis-33630	8	18	)	)	PUNCT
bracis-33630	8	19	as	as	ADP
bracis-33630	8	20	an	an	DET
bracis-33630	8	21	alternative	alternative	NOUN
bracis-33630	8	22	metric	metric	ADJ
bracis-33630	8	23	to	to	PART
bracis-33630	8	24	reduce	reduce	VERB
bracis-33630	8	25	the	the	DET
bracis-33630	8	26	formation	formation	NOUN
bracis-33630	8	27	of	of	ADP
bracis-33630	8	28	single	single	ADJ
bracis-33630	8	29	-	-	PUNCT
bracis-33630	8	30	element	element	NOUN
bracis-33630	8	31	clusters	cluster	NOUN
bracis-33630	8	32	and	and	CCONJ
bracis-33630	8	33	provide	provide	VERB
bracis-33630	8	34	a	a	DET
bracis-33630	8	35	more	more	ADV
bracis-33630	8	36	balanced	balanced	ADJ
bracis-33630	8	37	measure	measure	NOUN
bracis-33630	8	38	of	of	ADP
bracis-33630	8	39	clustering	clustering	ADJ
bracis-33630	8	40	performance	performance	NOUN
bracis-33630	8	41	.	.	PUNCT
bracis-33630	9	1	we	we	PRON
bracis-33630	9	2	present	present	VERB
bracis-33630	9	3	new	new	ADJ
bracis-33630	9	4	insights	insight	NOUN
bracis-33630	9	5	for	for	ADP
bracis-33630	9	6	applying	apply	VERB
bracis-33630	9	7	short	short	ADJ
bracis-33630	9	8	text	text	NOUN
bracis-33630	9	9	clustering	clustering	NOUN
bracis-33630	9	10	methodologies	methodology	NOUN
bracis-33630	9	11	in	in	ADP
bracis-33630	9	12	real	real	ADJ
bracis-33630	9	13	-	-	PUNCT
bracis-33630	9	14	world	world	NOUN
bracis-33630	9	15	situations	situation	NOUN
bracis-33630	9	16	,	,	PUNCT
bracis-33630	9	17	especially	especially	ADV
bracis-33630	9	18	in	in	ADP
bracis-33630	9	19	sectors	sector	NOUN
bracis-33630	9	20	like	like	ADP
bracis-33630	9	21	e	e	NOUN
bracis-33630	9	22	-	-	NOUN
bracis-33630	9	23	commerce	commerce	NOUN
bracis-33630	9	24	,	,	PUNCT
bracis-33630	9	25	where	where	SCONJ
bracis-33630	9	26	text	text	NOUN
bracis-33630	9	27	data	datum	NOUN
bracis-33630	9	28	volumes	volume	NOUN
bracis-33630	9	29	and	and	CCONJ
bracis-33630	9	30	dynamics	dynamic	NOUN
bracis-33630	9	31	present	present	VERB
bracis-33630	9	32	unique	unique	ADJ
bracis-33630	9	33	challenges	challenge	NOUN
bracis-33630	9	34	.	.	PUNCT
bracis-33630	10	1	access	access	NOUN
bracis-33630	10	2	provided	provide	VERB
bracis-33630	10	3	by	by	ADP
bracis-33630	10	4	university	university	PROPN
bracis-33630	10	5	of	of	ADP
bracis-33630	10	6	notre	notre	PROPN
bracis-33630	10	7	dame	dame	PROPN
bracis-33630	10	8	hesburgh	hesburgh	PROPN
bracis-33630	10	9	library	library	PROPN
bracis-33630	10	10	.	.	PUNCT
bracis-33630	11	1	download	download	PROPN
bracis-33630	11	2	conference	conference	NOUN
bracis-33630	11	3	paper	paper	NOUN
bracis-33630	11	4	pdf	pdf	NOUN
bracis-33630	11	5	similar	similar	ADJ
bracis-33630	11	6	content	content	NOUN
bracis-33630	11	7	being	be	AUX
bracis-33630	11	8	viewed	view	VERB
bracis-33630	11	9	by	by	ADP
bracis-33630	11	10	others	other	NOUN
bracis-33630	11	11	dynamic	dynamic	ADJ
bracis-33630	11	12	clustering	clustering	NOUN
bracis-33630	11	13	for	for	ADP
bracis-33630	11	14	short	short	ADJ
bracis-33630	11	15	text	text	NOUN
bracis-33630	11	16	stream	stream	NOUN
bracis-33630	11	17	based	base	VERB
bracis-33630	11	18	on	on	ADP
bracis-33630	11	19	dirichlet	dirichlet	PROPN
bracis-33630	11	20	process	process	NOUN
bracis-33630	11	21	article	article	NOUN
bracis-33630	11	22	26	26	NUM
bracis-33630	11	23	july	july	PROPN
bracis-33630	11	24	2021	2021	NUM
bracis-33630	11	25	topic	topic	NOUN
bracis-33630	11	26	model	model	NOUN
bracis-33630	11	27	with	with	ADP
bracis-33630	11	28	contextual	contextual	ADJ
bracis-33630	11	29	outlier	outlier	NOUN
bracis-33630	11	30	handling	handling	NOUN
bracis-33630	11	31	:	:	PUNCT
bracis-33630	11	32	a	a	DET
bracis-33630	11	33	study	study	NOUN
bracis-33630	11	34	on	on	ADP
bracis-33630	11	35	electronic	electronic	ADJ
bracis-33630	11	36	invoice	invoice	NOUN
bracis-33630	11	37	product	product	NOUN
bracis-33630	11	38	descriptions	description	NOUN
bracis-33630	11	39	chapter	chapter	NOUN
bracis-33630	11	40	©	©	PROPN
bracis-33630	11	41	2023	2023	NUM
bracis-33630	11	42	short	short	ADJ
bracis-33630	11	43	text	text	NOUN
bracis-33630	11	44	processing	processing	NOUN
bracis-33630	11	45	for	for	ADP
bracis-33630	11	46	analyzing	analyze	VERB
bracis-33630	11	47	user	user	NOUN
bracis-33630	11	48	portraits	portrait	NOUN
bracis-33630	11	49	:	:	PUNCT
bracis-33630	11	50	a	a	DET
bracis-33630	11	51	dynamic	dynamic	ADJ
bracis-33630	11	52	combination	combination	NOUN
bracis-33630	11	53	chapter	chapter	NOUN
bracis-33630	11	54	©	©	NOUN
bracis-33630	11	55	2020	2020	NUM
bracis-33630	11	56	1	1	NUM
bracis-33630	11	57	introduction	introduction	NOUN
bracis-33630	11	58	text	text	NOUN
bracis-33630	11	59	clustering	clustering	NOUN
bracis-33630	11	60	has	have	AUX
bracis-33630	11	61	become	become	VERB
bracis-33630	11	62	an	an	DET
bracis-33630	11	63	essential	essential	ADJ
bracis-33630	11	64	technique	technique	NOUN
bracis-33630	11	65	in	in	ADP
bracis-33630	11	66	natural	natural	ADJ
bracis-33630	11	67	language	language	NOUN
bracis-33630	11	68	processing	processing	NOUN
bracis-33630	11	69	(	(	PUNCT
bracis-33630	11	70	nlp	nlp	NOUN
bracis-33630	11	71	)	)	PUNCT
bracis-33630	11	72	,	,	PUNCT
bracis-33630	11	73	enabling	enable	VERB
bracis-33630	11	74	the	the	DET
bracis-33630	11	75	discovery	discovery	NOUN
bracis-33630	11	76	of	of	ADP
bracis-33630	11	77	hidden	hide	VERB
bracis-33630	11	78	structures	structure	NOUN
bracis-33630	11	79	within	within	ADP
bracis-33630	11	80	large	large	ADJ
bracis-33630	11	81	collections	collection	NOUN
bracis-33630	11	82	of	of	ADP
bracis-33630	11	83	unstructured	unstructured	ADJ
bracis-33630	11	84	text	text	NOUN
bracis-33630	11	85	data	datum	NOUN
bracis-33630	11	86	.	.	PUNCT
bracis-33630	12	1	among	among	ADP
bracis-33630	12	2	various	various	ADJ
bracis-33630	12	3	approaches	approach	NOUN
bracis-33630	12	4	,	,	PUNCT
bracis-33630	12	5	stream	stream	VERB
bracis-33630	12	6	short	short	ADJ
bracis-33630	12	7	text	text	NOUN
bracis-33630	12	8	clustering	clustering	NOUN
bracis-33630	12	9	,	,	PUNCT
bracis-33630	12	10	particularly	particularly	ADV
bracis-33630	12	11	methods	method	NOUN
bracis-33630	12	12	based	base	VERB
bracis-33630	12	13	on	on	ADP
bracis-33630	12	14	latent	latent	ADJ
bracis-33630	12	15	dirichlet	dirichlet	PROPN
bracis-33630	12	16	allocation	allocation	NOUN
bracis-33630	12	17	(	(	PUNCT
bracis-33630	12	18	lda	lda	PROPN
bracis-33630	12	19	)	)	PUNCT
bracis-33630	12	20	,	,	PUNCT
bracis-33630	12	21	has	have	AUX
bracis-33630	12	22	been	be	AUX
bracis-33630	12	23	widely	widely	ADV
bracis-33630	12	24	researched	research	VERB
bracis-33630	12	25	due	due	ADP
bracis-33630	12	26	to	to	ADP
bracis-33630	12	27	its	its	PRON
bracis-33630	12	28	effectiveness	effectiveness	NOUN
bracis-33630	12	29	in	in	ADP
bracis-33630	12	30	grouping	group	VERB
bracis-33630	12	31	similar	similar	ADJ
bracis-33630	12	32	texts	text	NOUN
bracis-33630	12	33	and	and	CCONJ
bracis-33630	12	34	discovering	discover	VERB
bracis-33630	12	35	topical	topical	ADJ
bracis-33630	12	36	patterns	pattern	NOUN
bracis-33630	12	37	.	.	PUNCT
bracis-33630	13	1	while	while	SCONJ
bracis-33630	13	2	these	these	DET
bracis-33630	13	3	methods	method	NOUN
bracis-33630	13	4	have	have	AUX
bracis-33630	13	5	shown	show	VERB
bracis-33630	13	6	promising	promising	ADJ
bracis-33630	13	7	results	result	NOUN
bracis-33630	13	8	in	in	ADP
bracis-33630	13	9	domains	domain	NOUN
bracis-33630	13	10	such	such	ADJ
bracis-33630	13	11	as	as	ADP
bracis-33630	13	12	news	news	NOUN
bracis-33630	13	13	feeds	feed	NOUN
bracis-33630	13	14	and	and	CCONJ
bracis-33630	13	15	social	social	ADJ
bracis-33630	13	16	media	medium	NOUN
bracis-33630	13	17	platforms	platform	NOUN
bracis-33630	13	18	,	,	PUNCT
bracis-33630	13	19	their	their	PRON
bracis-33630	13	20	application	application	NOUN
bracis-33630	13	21	to	to	ADP
bracis-33630	13	22	larger	large	ADJ
bracis-33630	13	23	and	and	CCONJ
bracis-33630	13	24	more	more	ADV
bracis-33630	13	25	complex	complex	ADJ
bracis-33630	13	26	datasets	dataset	NOUN
bracis-33630	13	27	remains	remain	VERB
bracis-33630	13	28	less	less	ADV
bracis-33630	13	29	explored	explore	VERB
bracis-33630	13	30	.	.	PUNCT
bracis-33630	14	1	the	the	DET
bracis-33630	14	2	performance	performance	NOUN
bracis-33630	14	3	of	of	ADP
bracis-33630	14	4	traditional	traditional	ADJ
bracis-33630	14	5	lda	lda	PROPN
bracis-33630	14	6	-	-	PUNCT
bracis-33630	14	7	based	base	VERB
bracis-33630	14	8	stream	stream	NOUN
bracis-33630	14	9	short	short	ADJ
bracis-33630	14	10	text	text	NOUN
bracis-33630	14	11	clustering	clustering	NOUN
bracis-33630	14	12	methods	method	NOUN
bracis-33630	14	13	is	be	AUX
bracis-33630	14	14	well	well	ADV
bracis-33630	14	15	-	-	PUNCT
bracis-33630	14	16	documented	document	VERB
bracis-33630	14	17	in	in	ADP
bracis-33630	14	18	scenarios	scenario	NOUN
bracis-33630	14	19	involving	involve	VERB
bracis-33630	14	20	relatively	relatively	ADV
bracis-33630	14	21	small	small	ADJ
bracis-33630	14	22	datasets	dataset	NOUN
bracis-33630	14	23	,	,	PUNCT
bracis-33630	14	24	with	with	ADP
bracis-33630	14	25	few	few	ADJ
bracis-33630	14	26	exceptions	exception	NOUN
bracis-33630	14	27	[	[	X
bracis-33630	14	28	1	1	NUM
bracis-33630	14	29	]	]	PUNCT
bracis-33630	14	30	.	.	PUNCT
bracis-33630	15	1	however	however	ADV
bracis-33630	15	2	,	,	PUNCT
bracis-33630	15	3	these	these	DET
bracis-33630	15	4	methods	method	NOUN
bracis-33630	15	5	often	often	ADV
bracis-33630	15	6	struggle	struggle	VERB
bracis-33630	15	7	when	when	SCONJ
bracis-33630	15	8	scaled	scale	VERB
bracis-33630	15	9	to	to	ADP
bracis-33630	15	10	larger	large	ADJ
bracis-33630	15	11	datasets	dataset	NOUN
bracis-33630	15	12	commonly	commonly	ADV
bracis-33630	15	13	found	find	VERB
bracis-33630	15	14	in	in	ADP
bracis-33630	15	15	industries	industry	NOUN
bracis-33630	15	16	like	like	ADP
bracis-33630	15	17	e	e	NOUN
bracis-33630	15	18	-	-	NOUN
bracis-33630	15	19	commerce	commerce	NOUN
bracis-33630	15	20	.	.	PUNCT
bracis-33630	16	1	the	the	DET
bracis-33630	16	2	dynamic	dynamic	ADJ
bracis-33630	16	3	and	and	CCONJ
bracis-33630	16	4	voluminous	voluminous	ADJ
bracis-33630	16	5	nature	nature	NOUN
bracis-33630	16	6	of	of	ADP
bracis-33630	16	7	e	e	NOUN
bracis-33630	16	8	-	-	NOUN
bracis-33630	16	9	commerce	commerce	NOUN
bracis-33630	16	10	text	text	NOUN
bracis-33630	16	11	data	datum	NOUN
bracis-33630	16	12	presents	present	VERB
bracis-33630	16	13	unique	unique	ADJ
bracis-33630	16	14	challenges	challenge	NOUN
bracis-33630	16	15	not	not	PART
bracis-33630	16	16	typically	typically	ADV
bracis-33630	16	17	encountered	encounter	VERB
bracis-33630	16	18	in	in	ADP
bracis-33630	16	19	other	other	ADJ
bracis-33630	16	20	domains	domain	NOUN
bracis-33630	16	21	such	such	ADJ
bracis-33630	16	22	as	as	ADP
bracis-33630	16	23	news	news	NOUN
bracis-33630	16	24	or	or	CCONJ
bracis-33630	16	25	social	social	ADJ
bracis-33630	16	26	media	medium	NOUN
bracis-33630	16	27	.	.	PUNCT
bracis-33630	17	1	handling	handle	VERB
bracis-33630	17	2	the	the	DET
bracis-33630	17	3	vast	vast	ADJ
bracis-33630	17	4	number	number	NOUN
bracis-33630	17	5	of	of	ADP
bracis-33630	17	6	product	product	NOUN
bracis-33630	17	7	descriptions	description	NOUN
bracis-33630	17	8	,	,	PUNCT
bracis-33630	17	9	user	user	NOUN
bracis-33630	17	10	reviews	review	NOUN
bracis-33630	17	11	,	,	PUNCT
bracis-33630	17	12	and	and	CCONJ
bracis-33630	17	13	query	query	NOUN
bracis-33630	17	14	logs	log	NOUN
bracis-33630	17	15	requires	require	VERB
bracis-33630	17	16	clustering	cluster	VERB
bracis-33630	17	17	algorithms	algorithm	NOUN
bracis-33630	17	18	that	that	PRON
bracis-33630	17	19	are	be	AUX
bracis-33630	17	20	robust	robust	ADJ
bracis-33630	17	21	,	,	PUNCT
bracis-33630	17	22	scalable	scalable	ADJ
bracis-33630	17	23	,	,	PUNCT
bracis-33630	17	24	and	and	CCONJ
bracis-33630	17	25	adaptive	adaptive	ADJ
bracis-33630	17	26	to	to	ADP
bracis-33630	17	27	the	the	DET
bracis-33630	17	28	evolving	evolve	VERB
bracis-33630	17	29	nature	nature	NOUN
bracis-33630	17	30	of	of	ADP
bracis-33630	17	31	data	data	PROPN
bracis-33630	17	32	.	.	PUNCT
bracis-33630	18	1	a	a	DET
bracis-33630	18	2	significant	significant	ADJ
bracis-33630	18	3	challenge	challenge	NOUN
bracis-33630	18	4	in	in	ADP
bracis-33630	18	5	evaluating	evaluate	VERB
bracis-33630	18	6	the	the	DET
bracis-33630	18	7	performance	performance	NOUN
bracis-33630	18	8	of	of	ADP
bracis-33630	18	9	clustering	clustering	ADJ
bracis-33630	18	10	methods	method	NOUN
bracis-33630	18	11	is	be	AUX
bracis-33630	18	12	the	the	DET
bracis-33630	18	13	reliance	reliance	NOUN
bracis-33630	18	14	on	on	ADP
bracis-33630	18	15	metrics	metric	NOUN
bracis-33630	18	16	like	like	ADP
bracis-33630	18	17	normalized	normalize	VERB
bracis-33630	18	18	mutual	mutual	ADJ
bracis-33630	18	19	information	information	NOUN
bracis-33630	18	20	(	(	PUNCT
bracis-33630	18	21	nmi	nmi	PROPN
bracis-33630	18	22	)	)	PUNCT
bracis-33630	18	23	.	.	PUNCT
bracis-33630	19	1	while	while	SCONJ
bracis-33630	19	2	nmi	nmi	PROPN
bracis-33630	19	3	is	be	AUX
bracis-33630	19	4	useful	useful	ADJ
bracis-33630	19	5	for	for	ADP
bracis-33630	19	6	measuring	measure	VERB
bracis-33630	19	7	the	the	DET
bracis-33630	19	8	statistical	statistical	ADJ
bracis-33630	19	9	accuracy	accuracy	NOUN
bracis-33630	19	10	of	of	ADP
bracis-33630	19	11	clustering	clustering	ADJ
bracis-33630	19	12	results	result	NOUN
bracis-33630	19	13	,	,	PUNCT
bracis-33630	19	14	it	it	PRON
bracis-33630	19	15	does	do	AUX
bracis-33630	19	16	not	not	PART
bracis-33630	19	17	fully	fully	ADV
bracis-33630	19	18	capture	capture	VERB
bracis-33630	19	19	the	the	DET
bracis-33630	19	20	practical	practical	ADJ
bracis-33630	19	21	aspects	aspect	NOUN
bracis-33630	19	22	of	of	ADP
bracis-33630	19	23	the	the	DET
bracis-33630	19	24	clusters	cluster	NOUN
bracis-33630	19	25	formed	form	VERB
bracis-33630	19	26	,	,	PUNCT
bracis-33630	19	27	particularly	particularly	ADV
bracis-33630	19	28	in	in	ADP
bracis-33630	19	29	the	the	DET
bracis-33630	19	30	presence	presence	NOUN
bracis-33630	19	31	of	of	ADP
bracis-33630	19	32	single	single	ADJ
bracis-33630	19	33	-	-	PUNCT
bracis-33630	19	34	element	element	NOUN
bracis-33630	19	35	clusters	cluster	NOUN
bracis-33630	19	36	,	,	PUNCT
bracis-33630	19	37	which	which	PRON
bracis-33630	19	38	can	can	AUX
bracis-33630	19	39	artificially	artificially	ADV
bracis-33630	19	40	inflate	inflate	VERB
bracis-33630	19	41	homogeneity	homogeneity	NOUN
bracis-33630	19	42	scores	score	NOUN
bracis-33630	19	43	[	[	X
bracis-33630	19	44	2	2	NUM
bracis-33630	19	45	,	,	PUNCT
bracis-33630	19	46	3	3	NUM
bracis-33630	19	47	]	]	PUNCT
bracis-33630	19	48	.	.	PUNCT
bracis-33630	20	1	this	this	DET
bracis-33630	20	2	limitation	limitation	NOUN
bracis-33630	20	3	calls	call	VERB
bracis-33630	20	4	for	for	SCONJ
bracis-33630	20	5	a	a	DET
bracis-33630	20	6	more	more	ADV
bracis-33630	20	7	comprehensive	comprehensive	ADJ
bracis-33630	20	8	evaluation	evaluation	NOUN
bracis-33630	20	9	framework	framework	NOUN
bracis-33630	20	10	to	to	PART
bracis-33630	20	11	better	well	ADV
bracis-33630	20	12	understand	understand	VERB
bracis-33630	20	13	the	the	DET
bracis-33630	20	14	practical	practical	ADJ
bracis-33630	20	15	utility	utility	NOUN
bracis-33630	20	16	of	of	ADP
bracis-33630	20	17	clustering	cluster	VERB
bracis-33630	20	18	outcomes	outcome	NOUN
bracis-33630	20	19	.	.	PUNCT
bracis-33630	21	1	our	our	PRON
bracis-33630	21	2	study	study	NOUN
bracis-33630	21	3	seeks	seek	VERB
bracis-33630	21	4	to	to	PART
bracis-33630	21	5	extend	extend	VERB
bracis-33630	21	6	the	the	DET
bracis-33630	21	7	application	application	NOUN
bracis-33630	21	8	of	of	ADP
bracis-33630	21	9	these	these	DET
bracis-33630	21	10	clustering	clustering	ADJ
bracis-33630	21	11	methods	method	NOUN
bracis-33630	21	12	to	to	PART
bracis-33630	21	13	understand	understand	VERB
bracis-33630	21	14	their	their	PRON
bracis-33630	21	15	scalability	scalability	NOUN
bracis-33630	21	16	and	and	CCONJ
bracis-33630	21	17	adaptability	adaptability	NOUN
bracis-33630	21	18	better	well	ADV
bracis-33630	21	19	,	,	PUNCT
bracis-33630	21	20	particularly	particularly	ADV
bracis-33630	21	21	focusing	focus	VERB
bracis-33630	21	22	on	on	ADP
bracis-33630	21	23	:	:	PUNCT
bracis-33630	21	24	1	1	X
bracis-33630	21	25	.	.	X
bracis-33630	22	1	how	how	SCONJ
bracis-33630	22	2	these	these	DET
bracis-33630	22	3	methods	method	NOUN
bracis-33630	22	4	perform	perform	VERB
bracis-33630	22	5	with	with	ADP
bracis-33630	22	6	extensive	extensive	ADJ
bracis-33630	22	7	e	e	NOUN
bracis-33630	22	8	-	-	NOUN
bracis-33630	22	9	commerce	commerce	NOUN
bracis-33630	22	10	datasets	dataset	NOUN
bracis-33630	22	11	;	;	PUNCT
bracis-33630	23	1	2	2	X
bracis-33630	23	2	.	.	X
bracis-33630	23	3	the	the	DET
bracis-33630	23	4	clustering	clustering	ADJ
bracis-33630	23	5	quality	quality	NOUN
bracis-33630	23	6	metrics	metric	NOUN
bracis-33630	23	7	,	,	PUNCT
bracis-33630	23	8	especially	especially	ADV
bracis-33630	23	9	concerning	concern	VERB
bracis-33630	23	10	homogeneity	homogeneity	NOUN
bracis-33630	23	11	scores	score	NOUN
bracis-33630	23	12	and	and	CCONJ
bracis-33630	23	13	single	single	ADJ
bracis-33630	23	14	-	-	PUNCT
bracis-33630	23	15	element	element	NOUN
bracis-33630	23	16	cluster	cluster	NOUN
bracis-33630	23	17	formation	formation	NOUN
bracis-33630	23	18	;	;	PUNCT
bracis-33630	23	19	3	3	X
bracis-33630	23	20	.	.	X
bracis-33630	24	1	the	the	DET
bracis-33630	24	2	potential	potential	NOUN
bracis-33630	24	3	for	for	ADP
bracis-33630	24	4	alternative	alternative	ADJ
bracis-33630	24	5	metrics	metric	NOUN
bracis-33630	24	6	like	like	ADP
bracis-33630	24	7	the	the	DET
bracis-33630	24	8	akaike	akaike	ADJ
bracis-33630	24	9	information	information	NOUN
bracis-33630	24	10	criterion	criterion	NOUN
bracis-33630	24	11	(	(	PUNCT
bracis-33630	24	12	aic	aic	PROPN
bracis-33630	24	13	)	)	PUNCT
bracis-33630	24	14	to	to	PART
bracis-33630	24	15	provide	provide	VERB
bracis-33630	24	16	more	more	ADJ
bracis-33630	24	17	reliable	reliable	ADJ
bracis-33630	24	18	evaluations	evaluation	NOUN
bracis-33630	24	19	of	of	ADP
bracis-33630	24	20	clustering	cluster	VERB
bracis-33630	24	21	outcomes	outcome	NOUN
bracis-33630	24	22	.	.	PUNCT
bracis-33630	25	1	this	this	DET
bracis-33630	25	2	paper	paper	NOUN
bracis-33630	25	3	is	be	AUX
bracis-33630	25	4	organized	organize	VERB
bracis-33630	25	5	as	as	SCONJ
bracis-33630	25	6	follows	follow	VERB
bracis-33630	25	7	.	.	PUNCT
bracis-33630	26	1	section	section	NOUN
bracis-33630	26	2	 	 	SPACE
bracis-33630	26	3	2	2	NUM
bracis-33630	26	4	reviews	review	NOUN
bracis-33630	26	5	related	relate	VERB
bracis-33630	26	6	work	work	NOUN
bracis-33630	26	7	on	on	ADP
bracis-33630	26	8	similarity	similarity	NOUN
bracis-33630	26	9	-	-	PUNCT
bracis-33630	26	10	based	base	VERB
bracis-33630	26	11	and	and	CCONJ
bracis-33630	26	12	model	model	NOUN
bracis-33630	26	13	-	-	PUNCT
bracis-33630	26	14	based	base	VERB
bracis-33630	26	15	stream	stream	NOUN
bracis-33630	26	16	clustering	cluster	VERB
bracis-33630	26	17	,	,	PUNCT
bracis-33630	26	18	detailing	detail	VERB
bracis-33630	26	19	the	the	DET
bracis-33630	26	20	online	online	ADJ
bracis-33630	26	21	approaches	approach	NOUN
bracis-33630	26	22	.	.	PUNCT
bracis-33630	27	1	section	section	NOUN
bracis-33630	27	2	 	 	SPACE
bracis-33630	27	3	3	3	NUM
bracis-33630	27	4	presents	present	VERB
bracis-33630	27	5	our	our	PRON
bracis-33630	27	6	proposal	proposal	NOUN
bracis-33630	27	7	and	and	CCONJ
bracis-33630	27	8	details	detail	NOUN
bracis-33630	27	9	the	the	DET
bracis-33630	27	10	dataset	dataset	NOUN
bracis-33630	27	11	,	,	PUNCT
bracis-33630	27	12	the	the	DET
bracis-33630	27	13	dataset	dataset	NOUN
bracis-33630	27	14	preparation	preparation	NOUN
bracis-33630	27	15	,	,	PUNCT
bracis-33630	27	16	the	the	DET
bracis-33630	27	17	evaluated	evaluated	ADJ
bracis-33630	27	18	methods	method	NOUN
bracis-33630	27	19	,	,	PUNCT
bracis-33630	27	20	and	and	CCONJ
bracis-33630	27	21	the	the	DET
bracis-33630	27	22	experimental	experimental	ADJ
bracis-33630	27	23	setup	setup	NOUN
bracis-33630	27	24	.	.	PUNCT
bracis-33630	28	1	section	section	NOUN
bracis-33630	28	2	 	 	SPACE
bracis-33630	28	3	4.2	4.2	NUM
bracis-33630	28	4	reports	report	VERB
bracis-33630	28	5	the	the	DET
bracis-33630	28	6	results	result	NOUN
bracis-33630	28	7	.	.	PUNCT
bracis-33630	29	1	section	section	NOUN
bracis-33630	29	2	 	 	SPACE
bracis-33630	29	3	5	5	NUM
bracis-33630	29	4	discusses	discuss	VERB
bracis-33630	29	5	various	various	ADJ
bracis-33630	29	6	aspects	aspect	NOUN
bracis-33630	29	7	of	of	ADP
bracis-33630	29	8	the	the	DET
bracis-33630	29	9	experiment	experiment	NOUN
bracis-33630	29	10	.	.	PUNCT
bracis-33630	30	1	finally	finally	ADV
bracis-33630	30	2	,	,	PUNCT
bracis-33630	30	3	sect	sect	NOUN
bracis-33630	30	4	.	.	PUNCT
bracis-33630	30	5	 	 	SPACE
bracis-33630	30	6	6	6	NUM
bracis-33630	30	7	concludes	conclude	VERB
bracis-33630	30	8	the	the	DET
bracis-33630	30	9	paper	paper	NOUN
bracis-33630	30	10	and	and	CCONJ
bracis-33630	30	11	suggests	suggest	VERB
bracis-33630	30	12	future	future	ADJ
bracis-33630	30	13	research	research	NOUN
bracis-33630	30	14	directions	direction	NOUN
bracis-33630	30	15	.	.	PUNCT
bracis-33630	31	1	2	2	NUM
bracis-33630	31	2	related	related	ADJ
bracis-33630	31	3	work	work	NOUN
bracis-33630	31	4	two	two	NUM
bracis-33630	31	5	prominent	prominent	ADJ
bracis-33630	31	6	areas	area	NOUN
bracis-33630	31	7	of	of	ADP
bracis-33630	31	8	study	study	NOUN
bracis-33630	31	9	emerge	emerge	VERB
bracis-33630	31	10	in	in	ADP
bracis-33630	31	11	clustering	cluster	VERB
bracis-33630	31	12	short	short	ADJ
bracis-33630	31	13	text	text	NOUN
bracis-33630	31	14	streams	stream	NOUN
bracis-33630	31	15	:	:	PUNCT
bracis-33630	31	16	similarity	similarity	NOUN
bracis-33630	31	17	-	-	PUNCT
bracis-33630	31	18	based	base	VERB
bracis-33630	31	19	and	and	CCONJ
bracis-33630	31	20	model	model	NOUN
bracis-33630	31	21	-	-	PUNCT
bracis-33630	31	22	based	base	VERB
bracis-33630	31	23	approaches	approach	NOUN
bracis-33630	31	24	.	.	PUNCT
bracis-33630	32	1	similarity	similarity	NOUN
bracis-33630	32	2	-	-	PUNCT
bracis-33630	32	3	based	base	VERB
bracis-33630	32	4	clustering	clustering	NOUN
bracis-33630	32	5	relies	rely	VERB
bracis-33630	32	6	on	on	ADP
bracis-33630	32	7	pairwise	pairwise	NOUN
bracis-33630	32	8	similarities	similarity	NOUN
bracis-33630	32	9	,	,	PUNCT
bracis-33630	32	10	while	while	SCONJ
bracis-33630	32	11	model	model	NOUN
bracis-33630	32	12	-	-	PUNCT
bracis-33630	32	13	based	base	VERB
bracis-33630	32	14	clustering	clustering	NOUN
bracis-33630	32	15	relies	rely	VERB
bracis-33630	32	16	on	on	ADP
bracis-33630	32	17	statistical	statistical	ADJ
bracis-33630	32	18	models	model	NOUN
bracis-33630	32	19	that	that	PRON
bracis-33630	32	20	define	define	VERB
bracis-33630	32	21	data	datum	NOUN
bracis-33630	32	22	distribution	distribution	NOUN
bracis-33630	32	23	.	.	PUNCT
bracis-33630	33	1	2.1	2.1	NUM
bracis-33630	33	2	similarity	similarity	NOUN
bracis-33630	33	3	-	-	PUNCT
bracis-33630	33	4	based	base	VERB
bracis-33630	33	5	clustering	clustering	ADJ
bracis-33630	33	6	recent	recent	ADJ
bracis-33630	33	7	advances	advance	NOUN
bracis-33630	33	8	,	,	PUNCT
bracis-33630	33	9	incorporating	incorporate	VERB
bracis-33630	33	10	pre	pre	ADJ
bracis-33630	33	11	-	-	ADJ
bracis-33630	33	12	trained	train	VERB
bracis-33630	33	13	language	language	NOUN
bracis-33630	33	14	models	model	NOUN
bracis-33630	33	15	like	like	ADP
bracis-33630	33	16	bert	bert	PROPN
bracis-33630	33	17	 	 	SPACE
bracis-33630	33	18	[	[	X
bracis-33630	33	19	4	4	X
bracis-33630	33	20	]	]	PUNCT
bracis-33630	33	21	coupled	couple	VERB
bracis-33630	33	22	with	with	ADP
bracis-33630	33	23	clustering	clustering	ADJ
bracis-33630	33	24	algorithms	algorithm	NOUN
bracis-33630	33	25	such	such	ADJ
bracis-33630	33	26	as	as	ADP
bracis-33630	33	27	hdbscan	hdbscan	PROPN
bracis-33630	33	28	 	 	SPACE
bracis-33630	33	29	[	[	X
bracis-33630	33	30	5	5	NUM
bracis-33630	33	31	]	]	PUNCT
bracis-33630	33	32	,	,	PUNCT
bracis-33630	33	33	have	have	AUX
bracis-33630	33	34	showcased	showcase	VERB
bracis-33630	33	35	as	as	ADP
bracis-33630	33	36	promising	promise	VERB
bracis-33630	33	37	in	in	ADP
bracis-33630	33	38	text	text	NOUN
bracis-33630	33	39	clustering	clustering	NOUN
bracis-33630	33	40	tasks	task	NOUN
bracis-33630	33	41	.	.	PUNCT
bracis-33630	34	1	bert	bert	PROPN
bracis-33630	34	2	generates	generate	VERB
bracis-33630	34	3	contextual	contextual	ADJ
bracis-33630	34	4	embeddings	embedding	NOUN
bracis-33630	34	5	,	,	PUNCT
bracis-33630	34	6	allowing	allow	VERB
bracis-33630	34	7	deeper	deep	ADJ
bracis-33630	34	8	context	context	NOUN
bracis-33630	34	9	understanding	understanding	NOUN
bracis-33630	34	10	and	and	CCONJ
bracis-33630	34	11	language	language	NOUN
bracis-33630	34	12	structure	structure	NOUN
bracis-33630	34	13	modeling	modeling	NOUN
bracis-33630	34	14	 	 	SPACE
bracis-33630	35	1	[	[	X
bracis-33630	35	2	6	6	NUM
bracis-33630	35	3	]	]	PUNCT
bracis-33630	35	4	.	.	PUNCT
bracis-33630	36	1	these	these	DET
bracis-33630	36	2	embeddings	embedding	NOUN
bracis-33630	36	3	serve	serve	VERB
bracis-33630	36	4	as	as	ADP
bracis-33630	36	5	inputs	input	NOUN
bracis-33630	36	6	for	for	ADP
bracis-33630	36	7	clustering	clustering	ADJ
bracis-33630	36	8	algorithms	algorithm	NOUN
bracis-33630	36	9	.	.	PUNCT
bracis-33630	37	1	studies	study	NOUN
bracis-33630	37	2	 	 	SPACE
bracis-33630	38	1	[	[	X
bracis-33630	38	2	7	7	NUM
bracis-33630	38	3	,	,	PUNCT
bracis-33630	38	4	8	8	NUM
bracis-33630	38	5	]	]	PUNCT
bracis-33630	38	6	have	have	AUX
bracis-33630	38	7	demonstrated	demonstrate	VERB
bracis-33630	38	8	the	the	DET
bracis-33630	38	9	superiority	superiority	NOUN
bracis-33630	38	10	of	of	ADP
bracis-33630	38	11	combining	combine	VERB
bracis-33630	38	12	bert	bert	NOUN
bracis-33630	38	13	embeddings	embedding	NOUN
bracis-33630	38	14	and	and	CCONJ
bracis-33630	38	15	hdbscan	hdbscan	NOUN
bracis-33630	38	16	for	for	ADP
bracis-33630	38	17	short	short	ADJ
bracis-33630	38	18	text	text	NOUN
bracis-33630	38	19	clustering	cluster	VERB
bracis-33630	38	20	over	over	ADP
bracis-33630	38	21	traditional	traditional	ADJ
bracis-33630	38	22	methods	method	NOUN
bracis-33630	38	23	.	.	PUNCT
bracis-33630	39	1	additionally	additionally	ADV
bracis-33630	39	2	,	,	PUNCT
bracis-33630	39	3	elinac	elinac	ADJ
bracis-33630	39	4	 	 	SPACE
bracis-33630	40	1	[	[	X
bracis-33630	40	2	9	9	NUM
bracis-33630	40	3	]	]	PUNCT
bracis-33630	40	4	introduced	introduce	VERB
bracis-33630	40	5	an	an	DET
bracis-33630	40	6	auto	auto	NOUN
bracis-33630	40	7	-	-	PUNCT
bracis-33630	40	8	encoder	encoder	NOUN
bracis-33630	40	9	-	-	PUNCT
bracis-33630	40	10	based	base	VERB
bracis-33630	40	11	method	method	NOUN
bracis-33630	40	12	for	for	ADP
bracis-33630	40	13	clustering	cluster	VERB
bracis-33630	40	14	electronic	electronic	ADJ
bracis-33630	40	15	invoices	invoice	NOUN
bracis-33630	40	16	.	.	PUNCT
bracis-33630	41	1	despite	despite	SCONJ
bracis-33630	41	2	bert	bert	PROPN
bracis-33630	41	3	and	and	CCONJ
bracis-33630	41	4	transformers	transformer	NOUN
bracis-33630	41	5	being	be	AUX
bracis-33630	41	6	a	a	DET
bracis-33630	41	7	new	new	ADJ
bracis-33630	41	8	trend	trend	NOUN
bracis-33630	41	9	,	,	PUNCT
bracis-33630	41	10	their	their	PRON
bracis-33630	41	11	use	use	NOUN
bracis-33630	41	12	in	in	ADP
bracis-33630	41	13	short	short	ADJ
bracis-33630	41	14	text	text	NOUN
bracis-33630	41	15	stream	stream	NOUN
bracis-33630	41	16	clustering	cluster	VERB
bracis-33630	41	17	with	with	ADP
bracis-33630	41	18	lda	lda	PROPN
bracis-33630	41	19	methods	method	NOUN
bracis-33630	41	20	brings	bring	VERB
bracis-33630	41	21	an	an	DET
bracis-33630	41	22	additional	additional	ADJ
bracis-33630	41	23	computational	computational	ADJ
bracis-33630	41	24	cost	cost	NOUN
bracis-33630	41	25	 	 	SPACE
bracis-33630	42	1	[	[	X
bracis-33630	42	2	10	10	NUM
bracis-33630	42	3	]	]	PUNCT
bracis-33630	42	4	,	,	PUNCT
bracis-33630	42	5	posing	pose	VERB
bracis-33630	42	6	scale	scale	NOUN
bracis-33630	42	7	issues	issue	NOUN
bracis-33630	42	8	,	,	PUNCT
bracis-33630	42	9	particularly	particularly	ADV
bracis-33630	42	10	for	for	ADP
bracis-33630	42	11	large	large	ADJ
bracis-33630	42	12	datasets	dataset	NOUN
bracis-33630	42	13	.	.	PUNCT
bracis-33630	43	1	2.2	2.2	NUM
bracis-33630	43	2	model	model	NOUN
bracis-33630	43	3	-	-	PUNCT
bracis-33630	43	4	based	base	VERB
bracis-33630	43	5	clustering	clustering	ADJ
bracis-33630	43	6	approaches	approach	NOUN
bracis-33630	43	7	to	to	ADP
bracis-33630	43	8	model	model	NOUN
bracis-33630	43	9	-	-	PUNCT
bracis-33630	43	10	based	base	VERB
bracis-33630	43	11	clustering	clustering	NOUN
bracis-33630	43	12	of	of	ADP
bracis-33630	43	13	short	short	ADJ
bracis-33630	43	14	text	text	NOUN
bracis-33630	43	15	streams	stream	NOUN
bracis-33630	43	16	fall	fall	VERB
bracis-33630	43	17	into	into	ADP
bracis-33630	43	18	two	two	NUM
bracis-33630	43	19	categories	category	NOUN
bracis-33630	43	20	:	:	PUNCT
bracis-33630	43	21	batch	batch	NOUN
bracis-33630	43	22	methods	method	NOUN
bracis-33630	43	23	and	and	CCONJ
bracis-33630	43	24	online	online	ADJ
bracis-33630	43	25	methods	method	NOUN
bracis-33630	43	26	.	.	PUNCT
bracis-33630	44	1	batch	batch	NOUN
bracis-33630	44	2	methods	method	NOUN
bracis-33630	44	3	handle	handle	VERB
bracis-33630	44	4	data	datum	NOUN
bracis-33630	44	5	in	in	ADP
bracis-33630	44	6	separate	separate	ADJ
bracis-33630	44	7	blocks	block	NOUN
bracis-33630	44	8	,	,	PUNCT
bracis-33630	44	9	whereas	whereas	SCONJ
bracis-33630	44	10	online	online	ADJ
bracis-33630	44	11	methods	method	NOUN
bracis-33630	44	12	continuously	continuously	ADV
bracis-33630	44	13	process	process	VERB
bracis-33630	44	14	new	new	ADJ
bracis-33630	44	15	incoming	incoming	ADJ
bracis-33630	44	16	data	datum	NOUN
bracis-33630	44	17	.	.	PUNCT
bracis-33630	45	1	batch	batch	NOUN
bracis-33630	45	2	approaches	approach	NOUN
bracis-33630	45	3	.	.	PUNCT
bracis-33630	46	1	latent	latent	ADJ
bracis-33630	46	2	dirichlet	dirichlet	PROPN
bracis-33630	46	3	allocation	allocation	NOUN
bracis-33630	46	4	(	(	PUNCT
bracis-33630	46	5	lda	lda	PROPN
bracis-33630	46	6	)	)	PUNCT
bracis-33630	47	1	[	[	X
bracis-33630	47	2	11	11	NUM
bracis-33630	47	3	]	]	PUNCT
bracis-33630	47	4	is	be	AUX
bracis-33630	47	5	a	a	DET
bracis-33630	47	6	model	model	NOUN
bracis-33630	47	7	-	-	PUNCT
bracis-33630	47	8	based	base	VERB
bracis-33630	47	9	stream	stream	NOUN
bracis-33630	47	10	clustering	clustering	NOUN
bracis-33630	47	11	technique	technique	NOUN
bracis-33630	47	12	that	that	PRON
bracis-33630	47	13	has	have	AUX
bracis-33630	47	14	inspired	inspire	VERB
bracis-33630	47	15	numerous	numerous	ADJ
bracis-33630	47	16	extensions	extension	NOUN
bracis-33630	47	17	.	.	PUNCT
bracis-33630	48	1	these	these	DET
bracis-33630	48	2	extensions	extension	NOUN
bracis-33630	48	3	address	address	VERB
bracis-33630	48	4	challenges	challenge	NOUN
bracis-33630	48	5	related	relate	VERB
bracis-33630	48	6	to	to	ADP
bracis-33630	48	7	topic	topic	NOUN
bracis-33630	48	8	evolution	evolution	NOUN
bracis-33630	48	9	,	,	PUNCT
bracis-33630	48	10	semantic	semantic	ADJ
bracis-33630	48	11	representation	representation	NOUN
bracis-33630	48	12	,	,	PUNCT
bracis-33630	48	13	and	and	CCONJ
bracis-33630	48	14	the	the	DET
bracis-33630	48	15	dynamic	dynamic	ADJ
bracis-33630	48	16	nature	nature	NOUN
bracis-33630	48	17	of	of	ADP
bracis-33630	48	18	text	text	NOUN
bracis-33630	48	19	streams	stream	NOUN
bracis-33630	48	20	.	.	PUNCT
bracis-33630	49	1	initially	initially	ADV
bracis-33630	49	2	,	,	PUNCT
bracis-33630	49	3	models	model	NOUN
bracis-33630	49	4	like	like	ADP
bracis-33630	49	5	dct	dct	PROPN
bracis-33630	50	1	[	[	X
bracis-33630	50	2	12	12	NUM
bracis-33630	50	3	]	]	PUNCT
bracis-33630	50	4	aimed	aim	VERB
bracis-33630	50	5	to	to	PART
bracis-33630	50	6	simplify	simplify	VERB
bracis-33630	50	7	topic	topic	NOUN
bracis-33630	50	8	assignments	assignment	NOUN
bracis-33630	50	9	by	by	ADP
bracis-33630	50	10	assigning	assign	VERB
bracis-33630	50	11	a	a	DET
bracis-33630	50	12	single	single	ADJ
bracis-33630	50	13	topic	topic	NOUN
bracis-33630	50	14	to	to	ADP
bracis-33630	50	15	each	each	DET
bracis-33630	50	16	document	document	NOUN
bracis-33630	50	17	.	.	PUNCT
bracis-33630	51	1	while	while	SCONJ
bracis-33630	51	2	effective	effective	ADJ
bracis-33630	51	3	for	for	ADP
bracis-33630	51	4	short	short	ADJ
bracis-33630	51	5	texts	text	NOUN
bracis-33630	51	6	,	,	PUNCT
bracis-33630	51	7	this	this	DET
bracis-33630	51	8	approach	approach	NOUN
bracis-33630	51	9	lacked	lack	VERB
bracis-33630	51	10	adaptability	adaptability	NOUN
bracis-33630	51	11	to	to	PART
bracis-33630	51	12	handle	handle	VERB
bracis-33630	51	13	changing	change	VERB
bracis-33630	51	14	topic	topic	NOUN
bracis-33630	51	15	counts	count	NOUN
bracis-33630	51	16	within	within	ADP
bracis-33630	51	17	streams	stream	NOUN
bracis-33630	51	18	.	.	PUNCT
bracis-33630	52	1	mstream	mstream	PROPN
bracis-33630	52	2	 	 	SPACE
bracis-33630	53	1	[	[	X
bracis-33630	53	2	13	13	NUM
bracis-33630	53	3	]	]	PUNCT
bracis-33630	53	4	emerged	emerge	VERB
bracis-33630	53	5	in	in	ADP
bracis-33630	53	6	response	response	NOUN
bracis-33630	53	7	to	to	ADP
bracis-33630	53	8	this	this	DET
bracis-33630	53	9	challenge	challenge	NOUN
bracis-33630	53	10	,	,	PUNCT
bracis-33630	53	11	managing	manage	VERB
bracis-33630	53	12	topic	topic	NOUN
bracis-33630	53	13	counts	count	NOUN
bracis-33630	53	14	by	by	ADP
bracis-33630	53	15	discarding	discard	VERB
bracis-33630	53	16	outdated	outdated	ADJ
bracis-33630	53	17	batch	batch	NOUN
bracis-33630	53	18	documents	document	NOUN
bracis-33630	53	19	and	and	CCONJ
bracis-33630	53	20	adapting	adapt	VERB
bracis-33630	53	21	to	to	ADP
bracis-33630	53	22	evolving	evolve	VERB
bracis-33630	53	23	topics	topic	NOUN
bracis-33630	53	24	.	.	PUNCT
bracis-33630	54	1	however	however	ADV
bracis-33630	54	2	,	,	PUNCT
bracis-33630	54	3	mstream	mstream	PROPN
bracis-33630	54	4	’s	’s	PART
bracis-33630	54	5	reliance	reliance	NOUN
bracis-33630	54	6	on	on	ADP
bracis-33630	54	7	single	single	ADJ
bracis-33630	54	8	-	-	PUNCT
bracis-33630	54	9	term	term	NOUN
bracis-33630	54	10	document	document	NOUN
bracis-33630	54	11	representation	representation	NOUN
bracis-33630	54	12	limited	limit	VERB
bracis-33630	54	13	its	its	PRON
bracis-33630	54	14	ability	ability	NOUN
bracis-33630	54	15	to	to	PART
bracis-33630	54	16	navigate	navigate	VERB
bracis-33630	54	17	semantic	semantic	ADJ
bracis-33630	54	18	spaces	space	NOUN
bracis-33630	54	19	,	,	PUNCT
bracis-33630	54	20	impacting	impact	VERB
bracis-33630	54	21	cluster	cluster	NOUN
bracis-33630	54	22	purity	purity	NOUN
bracis-33630	54	23	.	.	PUNCT
bracis-33630	55	1	to	to	PART
bracis-33630	55	2	enhance	enhance	VERB
bracis-33630	55	3	semantic	semantic	ADJ
bracis-33630	55	4	representation	representation	NOUN
bracis-33630	55	5	,	,	PUNCT
bracis-33630	55	6	npmm	npmm	VERB
bracis-33630	55	7	 	 	SPACE
bracis-33630	55	8	[	[	X
bracis-33630	55	9	14	14	NUM
bracis-33630	55	10	]	]	X
bracis-33630	55	11	utilized	utilize	VERB
bracis-33630	55	12	pre	pre	ADJ
bracis-33630	55	13	-	-	ADJ
bracis-33630	55	14	learned	learned	ADJ
bracis-33630	55	15	word	word	NOUN
bracis-33630	55	16	embeddings	embedding	NOUN
bracis-33630	55	17	,	,	PUNCT
bracis-33630	55	18	advancing	advance	VERB
bracis-33630	55	19	text	text	NOUN
bracis-33630	55	20	semantics	semantic	NOUN
bracis-33630	55	21	understanding	understanding	NOUN
bracis-33630	55	22	.	.	PUNCT
bracis-33630	56	1	however	however	ADV
bracis-33630	56	2	,	,	PUNCT
bracis-33630	56	3	these	these	DET
bracis-33630	56	4	embeddings	embedding	NOUN
bracis-33630	56	5	’	'	PUNCT
bracis-33630	56	6	static	static	ADJ
bracis-33630	56	7	,	,	PUNCT
bracis-33630	56	8	language	language	NOUN
bracis-33630	56	9	-	-	PUNCT
bracis-33630	56	10	dependent	dependent	ADJ
bracis-33630	56	11	nature	nature	NOUN
bracis-33630	56	12	constrained	constrain	VERB
bracis-33630	56	13	their	their	PRON
bracis-33630	56	14	adaptability	adaptability	NOUN
bracis-33630	56	15	to	to	ADP
bracis-33630	56	16	evolving	evolve	VERB
bracis-33630	56	17	semantic	semantic	ADJ
bracis-33630	56	18	landscapes	landscape	NOUN
bracis-33630	56	19	.	.	PUNCT
bracis-33630	57	1	building	build	VERB
bracis-33630	57	2	on	on	ADP
bracis-33630	57	3	npmm	npmm	PROPN
bracis-33630	57	4	’s	’s	PART
bracis-33630	57	5	limitations	limitation	NOUN
bracis-33630	57	6	,	,	PUNCT
bracis-33630	57	7	dp	dp	PROPN
bracis-33630	57	8	-	-	PUNCT
bracis-33630	57	9	bmm	bmm	NOUN
bracis-33630	57	10	 	 	SPACE
bracis-33630	57	11	[	[	X
bracis-33630	57	12	15	15	NUM
bracis-33630	57	13	]	]	PUNCT
bracis-33630	57	14	introduced	introduce	VERB
bracis-33630	57	15	unordered	unordered	ADJ
bracis-33630	57	16	sequences	sequence	NOUN
bracis-33630	57	17	of	of	ADP
bracis-33630	57	18	bi	bi	NOUN
bracis-33630	57	19	-	-	NOUN
bracis-33630	57	20	terms	term	NOUN
bracis-33630	57	21	for	for	ADP
bracis-33630	57	22	semantic	semantic	ADJ
bracis-33630	57	23	representation	representation	NOUN
bracis-33630	57	24	,	,	PUNCT
bracis-33630	57	25	dynamically	dynamically	ADV
bracis-33630	57	26	detecting	detect	VERB
bracis-33630	57	27	topic	topic	NOUN
bracis-33630	57	28	counts	count	NOUN
bracis-33630	57	29	.	.	PUNCT
bracis-33630	58	1	nonetheless	nonetheless	ADV
bracis-33630	58	2	,	,	PUNCT
bracis-33630	58	3	this	this	DET
bracis-33630	58	4	emphasis	emphasis	NOUN
bracis-33630	58	5	on	on	ADP
bracis-33630	58	6	semantic	semantic	ADJ
bracis-33630	58	7	representation	representation	NOUN
bracis-33630	58	8	did	do	AUX
bracis-33630	58	9	not	not	PART
bracis-33630	58	10	directly	directly	ADV
bracis-33630	58	11	address	address	VERB
bracis-33630	58	12	challenges	challenge	NOUN
bracis-33630	58	13	posed	pose	VERB
bracis-33630	58	14	by	by	ADP
bracis-33630	58	15	sparse	sparse	ADJ
bracis-33630	58	16	,	,	PUNCT
bracis-33630	58	17	high	high	ADJ
bracis-33630	58	18	-	-	PUNCT
bracis-33630	58	19	dimensional	dimensional	ADJ
bracis-33630	58	20	data	datum	NOUN
bracis-33630	58	21	.	.	PUNCT
bracis-33630	59	1	the	the	DET
bracis-33630	59	2	introduction	introduction	NOUN
bracis-33630	59	3	of	of	ADP
bracis-33630	59	4	dcss	dcss	NOUN
bracis-33630	59	5	 	 	SPACE
bracis-33630	60	1	[	[	X
bracis-33630	60	2	16	16	NUM
bracis-33630	60	3	]	]	PUNCT
bracis-33630	60	4	and	and	CCONJ
bracis-33630	60	5	faststream	faststream	NOUN
bracis-33630	60	6	 	 	SPACE
bracis-33630	61	1	[	[	X
bracis-33630	61	2	17	17	NUM
bracis-33630	61	3	]	]	PUNCT
bracis-33630	61	4	marked	mark	VERB
bracis-33630	61	5	significant	significant	ADJ
bracis-33630	61	6	milestones	milestone	NOUN
bracis-33630	61	7	.	.	PUNCT
bracis-33630	62	1	dcss	dcss	NOUN
bracis-33630	62	2	,	,	PUNCT
bracis-33630	62	3	utilizing	utilize	VERB
bracis-33630	62	4	the	the	DET
bracis-33630	62	5	dirichlet	dirichlet	PROPN
bracis-33630	62	6	process	process	NOUN
bracis-33630	62	7	,	,	PUNCT
bracis-33630	62	8	demonstrated	demonstrate	VERB
bracis-33630	62	9	superior	superior	ADJ
bracis-33630	62	10	stability	stability	NOUN
bracis-33630	62	11	and	and	CCONJ
bracis-33630	62	12	adaptability	adaptability	NOUN
bracis-33630	62	13	by	by	ADP
bracis-33630	62	14	autonomously	autonomously	ADV
bracis-33630	62	15	learning	learn	VERB
bracis-33630	62	16	topic	topic	NOUN
bracis-33630	62	17	counts	count	NOUN
bracis-33630	62	18	and	and	CCONJ
bracis-33630	62	19	adjusting	adjust	VERB
bracis-33630	62	20	to	to	ADP
bracis-33630	62	21	topic	topic	NOUN
bracis-33630	62	22	drift	drift	NOUN
bracis-33630	62	23	.	.	PUNCT
bracis-33630	63	1	faststream	faststream	PROPN
bracis-33630	63	2	introduced	introduce	VERB
bracis-33630	63	3	efficiency	efficiency	NOUN
bracis-33630	63	4	and	and	CCONJ
bracis-33630	63	5	adaptability	adaptability	NOUN
bracis-33630	63	6	improvements	improvement	NOUN
bracis-33630	63	7	,	,	PUNCT
bracis-33630	63	8	employing	employ	VERB
bracis-33630	63	9	a	a	DET
bracis-33630	63	10	novel	novel	ADJ
bracis-33630	63	11	cluster	cluster	NOUN
bracis-33630	63	12	indexing	indexing	NOUN
bracis-33630	63	13	mechanism	mechanism	NOUN
bracis-33630	63	14	and	and	CCONJ
bracis-33630	63	15	dynamic	dynamic	ADJ
bracis-33630	63	16	similarity	similarity	NOUN
bracis-33630	63	17	thresholds	threshold	NOUN
bracis-33630	63	18	to	to	PART
bracis-33630	63	19	expedite	expedite	VERB
bracis-33630	63	20	processing	processing	NOUN
bracis-33630	63	21	.	.	PUNCT
bracis-33630	64	1	rakib	rakib	PROPN
bracis-33630	64	2	et	et	PROPN
bracis-33630	64	3	al	al	PROPN
bracis-33630	64	4	.	.	PUNCT
bracis-33630	64	5	 	 	SPACE
bracis-33630	65	1	[	[	X
bracis-33630	65	2	18	18	NUM
bracis-33630	65	3	]	]	PUNCT
bracis-33630	65	4	proposed	propose	VERB
bracis-33630	65	5	an	an	DET
bracis-33630	65	6	iterative	iterative	NOUN
bracis-33630	65	7	classification	classification	NOUN
bracis-33630	65	8	method	method	NOUN
bracis-33630	65	9	,	,	PUNCT
bracis-33630	65	10	enhancing	enhance	VERB
bracis-33630	65	11	short	short	ADJ
bracis-33630	65	12	text	text	NOUN
bracis-33630	65	13	clustering	clustering	NOUN
bracis-33630	65	14	.	.	PUNCT
bracis-33630	66	1	however	however	ADV
bracis-33630	66	2	,	,	PUNCT
bracis-33630	66	3	a	a	DET
bracis-33630	66	4	common	common	ADJ
bracis-33630	66	5	limitation	limitation	NOUN
bracis-33630	66	6	among	among	ADP
bracis-33630	66	7	these	these	DET
bracis-33630	66	8	studies	study	NOUN
bracis-33630	66	9	is	be	AUX
bracis-33630	66	10	that	that	SCONJ
bracis-33630	66	11	they	they	PRON
bracis-33630	66	12	are	be	AUX
bracis-33630	66	13	not	not	PART
bracis-33630	66	14	focused	focus	VERB
bracis-33630	66	15	on	on	ADP
bracis-33630	66	16	complex	complex	ADJ
bracis-33630	66	17	,	,	PUNCT
bracis-33630	66	18	large	large	ADJ
bracis-33630	66	19	datasets	dataset	NOUN
bracis-33630	66	20	,	,	PUNCT
bracis-33630	66	21	especially	especially	ADV
bracis-33630	66	22	in	in	ADP
bracis-33630	66	23	e	e	NOUN
bracis-33630	66	24	-	-	NOUN
bracis-33630	66	25	commerce	commerce	NOUN
bracis-33630	66	26	.	.	PUNCT
bracis-33630	67	1	online	online	ADJ
bracis-33630	67	2	approaches	approach	NOUN
bracis-33630	67	3	.	.	PUNCT
bracis-33630	68	1	the	the	DET
bracis-33630	68	2	online	online	ADJ
bracis-33630	68	3	semantic	semantic	ADV
bracis-33630	68	4	-	-	PUNCT
bracis-33630	68	5	enhanced	enhance	VERB
bracis-33630	68	6	dirichlet	dirichlet	PROPN
bracis-33630	68	7	model	model	NOUN
bracis-33630	68	8	(	(	PUNCT
bracis-33630	68	9	osdm	osdm	PROPN
bracis-33630	68	10	)	)	PUNCT
bracis-33630	68	11	[	[	X
bracis-33630	68	12	19	19	NUM
bracis-33630	68	13	]	]	PUNCT
bracis-33630	68	14	excels	excel	NOUN
bracis-33630	68	15	in	in	ADP
bracis-33630	68	16	dynamic	dynamic	ADJ
bracis-33630	68	17	clustering	clustering	NOUN
bracis-33630	68	18	,	,	PUNCT
bracis-33630	68	19	integrating	integrate	VERB
bracis-33630	68	20	word	word	NOUN
bracis-33630	68	21	-	-	PUNCT
bracis-33630	68	22	occurrence	occurrence	NOUN
bracis-33630	68	23	semantic	semantic	ADJ
bracis-33630	68	24	data	datum	NOUN
bracis-33630	68	25	into	into	ADP
bracis-33630	68	26	a	a	DET
bracis-33630	68	27	graphical	graphical	ADJ
bracis-33630	68	28	model	model	NOUN
bracis-33630	68	29	for	for	ADP
bracis-33630	68	30	real	real	ADJ
bracis-33630	68	31	-	-	PUNCT
bracis-33630	68	32	time	time	NOUN
bracis-33630	68	33	text	text	NOUN
bracis-33630	68	34	processing	processing	NOUN
bracis-33630	68	35	without	without	ADP
bracis-33630	68	36	fixed	fix	VERB
bracis-33630	68	37	cluster	cluster	NOUN
bracis-33630	68	38	numbers	number	NOUN
bracis-33630	68	39	or	or	CCONJ
bracis-33630	68	40	batch	batch	NOUN
bracis-33630	68	41	sizes	size	NOUN
bracis-33630	68	42	.	.	PUNCT
bracis-33630	68	43	tested	test	VERB
bracis-33630	68	44	on	on	ADP
bracis-33630	68	45	the	the	DET
bracis-33630	68	46	news	news	NOUN
bracis-33630	68	47	,	,	PUNCT
bracis-33630	68	48	reuters	reuter	NOUN
bracis-33630	68	49	,	,	PUNCT
bracis-33630	68	50	and	and	CCONJ
bracis-33630	68	51	tweets	tweet	NOUN
bracis-33630	68	52	datasets	dataset	NOUN
bracis-33630	68	53	and	and	CCONJ
bracis-33630	68	54	their	their	PRON
bracis-33630	68	55	synthetic	synthetic	ADJ
bracis-33630	68	56	versions	version	NOUN
bracis-33630	68	57	,	,	PUNCT
bracis-33630	68	58	osdm	osdm	PROPN
bracis-33630	68	59	uses	use	VERB
bracis-33630	68	60	metrics	metric	NOUN
bracis-33630	68	61	like	like	ADP
bracis-33630	68	62	purity	purity	NOUN
bracis-33630	68	63	,	,	PUNCT
bracis-33630	68	64	v	v	NOUN
bracis-33630	68	65	-	-	PUNCT
bracis-33630	68	66	measure	measure	NOUN
bracis-33630	68	67	,	,	PUNCT
bracis-33630	68	68	homogeneity	homogeneity	NOUN
bracis-33630	68	69	,	,	PUNCT
bracis-33630	68	70	and	and	CCONJ
bracis-33630	68	71	normalized	normalize	VERB
bracis-33630	68	72	mutual	mutual	ADJ
bracis-33630	68	73	information	information	NOUN
bracis-33630	68	74	(	(	PUNCT
bracis-33630	68	75	nmi	nmi	PROPN
bracis-33630	68	76	)	)	PUNCT
bracis-33630	68	77	for	for	ADP
bracis-33630	68	78	evaluation	evaluation	NOUN
bracis-33630	68	79	,	,	PUNCT
bracis-33630	68	80	adding	add	VERB
bracis-33630	68	81	accuracy	accuracy	NOUN
bracis-33630	68	82	to	to	ADP
bracis-33630	68	83	its	its	PRON
bracis-33630	68	84	metric	metric	ADJ
bracis-33630	68	85	suite	suite	NOUN
bracis-33630	68	86	for	for	ADP
bracis-33630	68	87	a	a	DET
bracis-33630	68	88	thorough	thorough	ADJ
bracis-33630	68	89	analysis	analysis	NOUN
bracis-33630	68	90	of	of	ADP
bracis-33630	68	91	performance	performance	NOUN
bracis-33630	68	92	.	.	PUNCT
bracis-33630	69	1	following	follow	VERB
bracis-33630	69	2	in	in	ADP
bracis-33630	69	3	the	the	DET
bracis-33630	69	4	footsteps	footstep	NOUN
bracis-33630	69	5	of	of	ADP
bracis-33630	69	6	osdm	osdm	NOUN
bracis-33630	69	7	,	,	PUNCT
bracis-33630	69	8	the	the	DET
bracis-33630	69	9	online	online	ADJ
bracis-33630	69	10	semantic	semantic	ADV
bracis-33630	69	11	-	-	PUNCT
bracis-33630	69	12	enhanced	enhance	VERB
bracis-33630	69	13	graphical	graphical	ADJ
bracis-33630	69	14	model	model	NOUN
bracis-33630	69	15	(	(	PUNCT
bracis-33630	69	16	osgm	osgm	NOUN
bracis-33630	69	17	)	)	PUNCT
bracis-33630	70	1	[	[	X
bracis-33630	70	2	20	20	NUM
bracis-33630	70	3	]	]	PUNCT
bracis-33630	70	4	also	also	ADV
bracis-33630	70	5	employs	employ	VERB
bracis-33630	70	6	a	a	DET
bracis-33630	70	7	poly	poly	ADJ
bracis-33630	70	8	urn	urn	NOUN
bracis-33630	70	9	scheme	scheme	NOUN
bracis-33630	70	10	to	to	PART
bracis-33630	70	11	dynamically	dynamically	ADV
bracis-33630	70	12	manage	manage	VERB
bracis-33630	70	13	document	document	NOUN
bracis-33630	70	14	clustering	clustering	NOUN
bracis-33630	70	15	,	,	PUNCT
bracis-33630	70	16	focusing	focus	VERB
bracis-33630	70	17	on	on	ADP
bracis-33630	70	18	semantic	semantic	ADJ
bracis-33630	70	19	smoothing	smoothing	NOUN
bracis-33630	70	20	and	and	CCONJ
bracis-33630	70	21	term	term	NOUN
bracis-33630	70	22	co	co	NOUN
bracis-33630	70	23	-	-	NOUN
bracis-33630	70	24	occurrence	occurrence	NOUN
bracis-33630	70	25	for	for	ADP
bracis-33630	70	26	enhanced	enhanced	ADJ
bracis-33630	70	27	clarity	clarity	NOUN
bracis-33630	70	28	and	and	CCONJ
bracis-33630	70	29	ambiguity	ambiguity	NOUN
bracis-33630	70	30	resolution	resolution	NOUN
bracis-33630	70	31	.	.	PUNCT
bracis-33630	71	1	osgm	osgm	PROPN
bracis-33630	71	2	’s	’s	PART
bracis-33630	71	3	evaluation	evaluation	NOUN
bracis-33630	71	4	mirrors	mirror	NOUN
bracis-33630	71	5	osdm	osdm	VERB
bracis-33630	71	6	’s	’s	PART
bracis-33630	71	7	,	,	PUNCT
bracis-33630	71	8	utilizing	utilize	VERB
bracis-33630	71	9	the	the	DET
bracis-33630	71	10	same	same	ADJ
bracis-33630	71	11	datasets	dataset	NOUN
bracis-33630	71	12	and	and	CCONJ
bracis-33630	71	13	metrics	metric	NOUN
bracis-33630	71	14	to	to	PART
bracis-33630	71	15	ensure	ensure	VERB
bracis-33630	71	16	consistency	consistency	NOUN
bracis-33630	71	17	in	in	ADP
bracis-33630	71	18	performance	performance	NOUN
bracis-33630	71	19	assessment	assessment	NOUN
bracis-33630	71	20	and	and	CCONJ
bracis-33630	71	21	comparability	comparability	NOUN
bracis-33630	71	22	between	between	ADP
bracis-33630	71	23	methods	method	NOUN
bracis-33630	71	24	.	.	PUNCT
bracis-33630	72	1	the	the	DET
bracis-33630	72	2	introduction	introduction	NOUN
bracis-33630	72	3	of	of	ADP
bracis-33630	72	4	eindm	eindm	PROPN
bracis-33630	72	5	[	[	X
bracis-33630	72	6	21	21	NUM
bracis-33630	72	7	]	]	X
bracis-33630	72	8	,	,	PUNCT
bracis-33630	72	9	a	a	DET
bracis-33630	72	10	context	context	NOUN
bracis-33630	72	11	-	-	PUNCT
bracis-33630	72	12	enhanced	enhance	VERB
bracis-33630	72	13	dirichlet	dirichlet	NOUN
bracis-33630	72	14	model	model	NOUN
bracis-33630	72	15	for	for	ADP
bracis-33630	72	16	online	online	ADJ
bracis-33630	72	17	clustering	clustering	NOUN
bracis-33630	72	18	in	in	ADP
bracis-33630	72	19	short	short	ADJ
bracis-33630	72	20	text	text	NOUN
bracis-33630	72	21	streams	stream	NOUN
bracis-33630	72	22	,	,	PUNCT
bracis-33630	72	23	adds	add	VERB
bracis-33630	72	24	another	another	DET
bracis-33630	72	25	layer	layer	NOUN
bracis-33630	72	26	to	to	ADP
bracis-33630	72	27	the	the	DET
bracis-33630	72	28	field	field	NOUN
bracis-33630	72	29	.	.	PUNCT
bracis-33630	73	1	by	by	ADP
bracis-33630	73	2	using	use	VERB
bracis-33630	73	3	a	a	DET
bracis-33630	73	4	window	window	NOUN
bracis-33630	73	5	-	-	PUNCT
bracis-33630	73	6	based	base	VERB
bracis-33630	73	7	semantic	semantic	ADJ
bracis-33630	73	8	term	term	NOUN
bracis-33630	73	9	representation	representation	NOUN
bracis-33630	73	10	and	and	CCONJ
bracis-33630	73	11	new	new	ADJ
bracis-33630	73	12	clustering	cluster	VERB
bracis-33630	73	13	metrics	metric	NOUN
bracis-33630	73	14	like	like	ADP
bracis-33630	73	15	word	word	NOUN
bracis-33630	73	16	specificity	specificity	NOUN
bracis-33630	73	17	,	,	PUNCT
bracis-33630	73	18	eindm	eindm	NOUN
bracis-33630	73	19	effectively	effectively	ADV
bracis-33630	73	20	captures	capture	VERB
bracis-33630	73	21	the	the	DET
bracis-33630	73	22	evolving	evolve	VERB
bracis-33630	73	23	nature	nature	NOUN
bracis-33630	73	24	of	of	ADP
bracis-33630	73	25	text	text	NOUN
bracis-33630	73	26	data	datum	NOUN
bracis-33630	73	27	.	.	PUNCT
bracis-33630	74	1	tested	test	VERB
bracis-33630	74	2	on	on	ADP
bracis-33630	74	3	datasets	dataset	NOUN
bracis-33630	74	4	similar	similar	ADJ
bracis-33630	74	5	to	to	ADP
bracis-33630	74	6	osdm	osdm	NOUN
bracis-33630	74	7	and	and	CCONJ
bracis-33630	74	8	osgm	osgm	NOUN
bracis-33630	74	9	,	,	PUNCT
bracis-33630	74	10	eindm	eindm	PROPN
bracis-33630	74	11	uses	use	VERB
bracis-33630	74	12	the	the	DET
bracis-33630	74	13	same	same	ADJ
bracis-33630	74	14	metrics	metric	NOUN
bracis-33630	74	15	except	except	SCONJ
bracis-33630	74	16	for	for	ADP
bracis-33630	74	17	accuracy	accuracy	NOUN
bracis-33630	74	18	,	,	PUNCT
bracis-33630	74	19	focusing	focus	VERB
bracis-33630	74	20	on	on	ADP
bracis-33630	74	21	core	core	NOUN
bracis-33630	74	22	aspects	aspect	NOUN
bracis-33630	74	23	of	of	ADP
bracis-33630	74	24	clustering	clustering	ADJ
bracis-33630	74	25	effectiveness	effectiveness	NOUN
bracis-33630	74	26	.	.	PUNCT
bracis-33630	75	1	estream	estream	NOUN
bracis-33630	76	1	[	[	X
bracis-33630	76	2	22	22	NUM
bracis-33630	76	3	]	]	PUNCT
bracis-33630	76	4	merges	merge	VERB
bracis-33630	76	5	online	online	ADV
bracis-33630	76	6	and	and	CCONJ
bracis-33630	76	7	offline	offline	ADJ
bracis-33630	76	8	clustering	clustering	NOUN
bracis-33630	76	9	techniques	technique	NOUN
bracis-33630	76	10	to	to	PART
bracis-33630	76	11	offer	offer	VERB
bracis-33630	76	12	a	a	DET
bracis-33630	76	13	versatile	versatile	ADJ
bracis-33630	76	14	solution	solution	NOUN
bracis-33630	76	15	to	to	ADP
bracis-33630	76	16	text	text	NOUN
bracis-33630	76	17	stream	stream	NOUN
bracis-33630	76	18	clustering	cluster	VERB
bracis-33630	76	19	.	.	PUNCT
bracis-33630	77	1	unlike	unlike	ADP
bracis-33630	77	2	the	the	DET
bracis-33630	77	3	other	other	ADJ
bracis-33630	77	4	methods	method	NOUN
bracis-33630	77	5	,	,	PUNCT
bracis-33630	77	6	estream	estream	NOUN
bracis-33630	77	7	expands	expand	VERB
bracis-33630	77	8	its	its	PRON
bracis-33630	77	9	testing	testing	NOUN
bracis-33630	77	10	grounds	ground	NOUN
bracis-33630	77	11	to	to	PART
bracis-33630	77	12	include	include	VERB
bracis-33630	77	13	additional	additional	ADJ
bracis-33630	77	14	datasets	dataset	NOUN
bracis-33630	77	15	like	like	ADP
bracis-33630	77	16	nt	not	PART
bracis-33630	77	17	and	and	CCONJ
bracis-33630	77	18	so	so	ADV
bracis-33630	77	19	-	-	PUNCT
bracis-33630	77	20	t	t	NOUN
bracis-33630	77	21	,	,	PUNCT
bracis-33630	77	22	along	along	ADP
bracis-33630	77	23	with	with	ADP
bracis-33630	77	24	news	news	NOUN
bracis-33630	77	25	-	-	PUNCT
bracis-33630	77	26	t	t	PROPN
bracis-33630	77	27	and	and	CCONJ
bracis-33630	77	28	tweets	tweet	NOUN
bracis-33630	77	29	-	-	PUNCT
bracis-33630	77	30	t.	t.	NOUN
bracis-33630	77	31	this	this	DET
bracis-33630	77	32	approach	approach	NOUN
bracis-33630	77	33	allows	allow	VERB
bracis-33630	77	34	for	for	ADP
bracis-33630	77	35	a	a	DET
bracis-33630	77	36	broader	broad	ADJ
bracis-33630	77	37	validation	validation	NOUN
bracis-33630	77	38	of	of	ADP
bracis-33630	77	39	its	its	PRON
bracis-33630	77	40	effectiveness	effectiveness	NOUN
bracis-33630	77	41	,	,	PUNCT
bracis-33630	77	42	using	use	VERB
bracis-33630	77	43	the	the	DET
bracis-33630	77	44	same	same	ADJ
bracis-33630	77	45	core	core	NOUN
bracis-33630	77	46	metrics	metric	NOUN
bracis-33630	77	47	to	to	PART
bracis-33630	77	48	ensure	ensure	VERB
bracis-33630	77	49	a	a	DET
bracis-33630	77	50	comprehensive	comprehensive	ADJ
bracis-33630	77	51	assessment	assessment	NOUN
bracis-33630	77	52	of	of	ADP
bracis-33630	77	53	its	its	PRON
bracis-33630	77	54	performance	performance	NOUN
bracis-33630	77	55	across	across	ADP
bracis-33630	77	56	various	various	ADJ
bracis-33630	77	57	text	text	NOUN
bracis-33630	77	58	stream	stream	NOUN
bracis-33630	77	59	scenarios	scenario	NOUN
bracis-33630	77	60	.	.	PUNCT
bracis-33630	78	1	while	while	SCONJ
bracis-33630	78	2	these	these	DET
bracis-33630	78	3	methods	method	NOUN
bracis-33630	78	4	in	in	ADP
bracis-33630	78	5	clustering	cluster	VERB
bracis-33630	78	6	short	short	ADJ
bracis-33630	78	7	text	text	NOUN
bracis-33630	78	8	streams	stream	NOUN
bracis-33630	78	9	show	show	VERB
bracis-33630	78	10	innovation	innovation	NOUN
bracis-33630	78	11	and	and	CCONJ
bracis-33630	78	12	effectiveness	effectiveness	NOUN
bracis-33630	78	13	,	,	PUNCT
bracis-33630	78	14	particularly	particularly	ADV
bracis-33630	78	15	when	when	SCONJ
bracis-33630	78	16	evaluated	evaluate	VERB
bracis-33630	78	17	with	with	ADP
bracis-33630	78	18	news	news	NOUN
bracis-33630	78	19	and	and	CCONJ
bracis-33630	78	20	tweets	tweet	NOUN
bracis-33630	78	21	datasets	dataset	NOUN
bracis-33630	78	22	.	.	PUNCT
bracis-33630	79	1	however	however	ADV
bracis-33630	79	2	,	,	PUNCT
bracis-33630	79	3	the	the	DET
bracis-33630	79	4	focus	focus	NOUN
bracis-33630	79	5	on	on	ADP
bracis-33630	79	6	these	these	DET
bracis-33630	79	7	limited	limit	VERB
bracis-33630	79	8	domains	domain	NOUN
bracis-33630	79	9	raises	raise	VERB
bracis-33630	79	10	questions	question	NOUN
bracis-33630	79	11	about	about	ADP
bracis-33630	79	12	their	their	PRON
bracis-33630	79	13	adaptability	adaptability	NOUN
bracis-33630	79	14	and	and	CCONJ
bracis-33630	79	15	performance	performance	NOUN
bracis-33630	79	16	in	in	ADP
bracis-33630	79	17	broader	broad	ADJ
bracis-33630	79	18	contexts	context	NOUN
bracis-33630	79	19	,	,	PUNCT
bracis-33630	79	20	such	such	ADJ
bracis-33630	79	21	as	as	ADP
bracis-33630	79	22	larger	large	ADJ
bracis-33630	79	23	datasets	dataset	NOUN
bracis-33630	79	24	or	or	CCONJ
bracis-33630	79	25	specific	specific	ADJ
bracis-33630	79	26	fields	field	NOUN
bracis-33630	79	27	like	like	ADP
bracis-33630	79	28	e	e	NOUN
bracis-33630	79	29	-	-	NOUN
bracis-33630	79	30	commerce	commerce	NOUN
bracis-33630	79	31	.	.	PUNCT
bracis-33630	79	32	exploring	explore	VERB
bracis-33630	79	33	varied	varied	ADJ
bracis-33630	79	34	datasets	dataset	NOUN
bracis-33630	79	35	could	could	AUX
bracis-33630	79	36	provide	provide	VERB
bracis-33630	79	37	deeper	deep	ADJ
bracis-33630	79	38	insights	insight	NOUN
bracis-33630	79	39	into	into	ADP
bracis-33630	79	40	the	the	DET
bracis-33630	79	41	scalability	scalability	NOUN
bracis-33630	79	42	and	and	CCONJ
bracis-33630	79	43	applicability	applicability	NOUN
bracis-33630	79	44	of	of	ADP
bracis-33630	79	45	these	these	DET
bracis-33630	79	46	clustering	clustering	ADJ
bracis-33630	79	47	techniques	technique	NOUN
bracis-33630	79	48	across	across	ADP
bracis-33630	79	49	diverse	diverse	ADJ
bracis-33630	79	50	real	real	ADJ
bracis-33630	79	51	-	-	PUNCT
bracis-33630	79	52	world	world	NOUN
bracis-33630	79	53	scenarios	scenario	NOUN
bracis-33630	79	54	.	.	PUNCT
bracis-33630	80	1	the	the	DET
bracis-33630	80	2	evaluation	evaluation	NOUN
bracis-33630	80	3	of	of	ADP
bracis-33630	80	4	clustering	clustering	ADJ
bracis-33630	80	5	methods	method	NOUN
bracis-33630	80	6	often	often	ADV
bracis-33630	80	7	relies	rely	VERB
bracis-33630	80	8	on	on	ADP
bracis-33630	80	9	normalized	normalize	VERB
bracis-33630	80	10	mutual	mutual	ADJ
bracis-33630	80	11	information	information	NOUN
bracis-33630	80	12	(	(	PUNCT
bracis-33630	80	13	nmi	nmi	PROPN
bracis-33630	80	14	)	)	PUNCT
bracis-33630	80	15	.	.	PUNCT
bracis-33630	81	1	however	however	ADV
bracis-33630	81	2	,	,	PUNCT
bracis-33630	81	3	nmi	nmi	PROPN
bracis-33630	81	4	alone	alone	ADV
bracis-33630	81	5	does	do	AUX
bracis-33630	81	6	not	not	PART
bracis-33630	81	7	account	account	VERB
bracis-33630	81	8	for	for	ADP
bracis-33630	81	9	qualitative	qualitative	ADJ
bracis-33630	81	10	factors	factor	NOUN
bracis-33630	81	11	like	like	ADP
bracis-33630	81	12	the	the	DET
bracis-33630	81	13	formation	formation	NOUN
bracis-33630	81	14	of	of	ADP
bracis-33630	81	15	very	very	ADV
bracis-33630	81	16	small	small	ADJ
bracis-33630	81	17	or	or	CCONJ
bracis-33630	81	18	single	single	ADJ
bracis-33630	81	19	-	-	PUNCT
bracis-33630	81	20	element	element	NOUN
bracis-33630	81	21	clusters	cluster	NOUN
bracis-33630	81	22	.	.	PUNCT
bracis-33630	82	1	a	a	DET
bracis-33630	82	2	more	more	ADV
bracis-33630	82	3	nuanced	nuanced	ADJ
bracis-33630	82	4	evaluation	evaluation	NOUN
bracis-33630	82	5	considering	consider	VERB
bracis-33630	82	6	cluster	cluster	NOUN
bracis-33630	82	7	size	size	NOUN
bracis-33630	82	8	distribution	distribution	NOUN
bracis-33630	82	9	and	and	CCONJ
bracis-33630	82	10	the	the	DET
bracis-33630	82	11	meaningfulness	meaningfulness	NOUN
bracis-33630	82	12	of	of	ADP
bracis-33630	82	13	clusters	cluster	NOUN
bracis-33630	82	14	formed	form	VERB
bracis-33630	82	15	could	could	AUX
bracis-33630	82	16	lead	lead	VERB
bracis-33630	82	17	to	to	ADP
bracis-33630	82	18	a	a	DET
bracis-33630	82	19	more	more	ADV
bracis-33630	82	20	comprehensive	comprehensive	ADJ
bracis-33630	82	21	understanding	understanding	NOUN
bracis-33630	82	22	of	of	ADP
bracis-33630	82	23	clustering	clustering	ADJ
bracis-33630	82	24	effectiveness	effectiveness	NOUN
bracis-33630	82	25	,	,	PUNCT
bracis-33630	82	26	ensuring	ensure	VERB
bracis-33630	82	27	statistical	statistical	ADJ
bracis-33630	82	28	performance	performance	NOUN
bracis-33630	82	29	and	and	CCONJ
bracis-33630	82	30	practical	practical	ADJ
bracis-33630	82	31	significance	significance	NOUN
bracis-33630	82	32	.	.	PUNCT
bracis-33630	83	1	3	3	NUM
bracis-33630	83	2	our	our	PRON
bracis-33630	83	3	methodology	methodology	NOUN
bracis-33630	83	4	recent	recent	ADJ
bracis-33630	83	5	studies	study	NOUN
bracis-33630	83	6	utilizing	utilize	VERB
bracis-33630	83	7	online	online	PROPN
bracis-33630	83	8	lda	lda	PROPN
bracis-33630	83	9	approaches	approach	VERB
bracis-33630	83	10	for	for	ADP
bracis-33630	83	11	clustering	cluster	VERB
bracis-33630	83	12	short	short	ADJ
bracis-33630	83	13	text	text	NOUN
bracis-33630	83	14	streams	stream	NOUN
bracis-33630	83	15	have	have	AUX
bracis-33630	83	16	reported	report	VERB
bracis-33630	83	17	substantial	substantial	ADJ
bracis-33630	83	18	performance	performance	NOUN
bracis-33630	83	19	achievements	achievement	NOUN
bracis-33630	83	20	.	.	PUNCT
bracis-33630	84	1	however	however	ADV
bracis-33630	84	2	,	,	PUNCT
bracis-33630	84	3	these	these	DET
bracis-33630	84	4	studies	study	NOUN
bracis-33630	84	5	have	have	AUX
bracis-33630	84	6	predominantly	predominantly	ADV
bracis-33630	84	7	focused	focus	VERB
bracis-33630	84	8	on	on	ADP
bracis-33630	84	9	smaller	small	ADJ
bracis-33630	84	10	datasets	dataset	NOUN
bracis-33630	84	11	,	,	PUNCT
bracis-33630	84	12	mainly	mainly	ADV
bracis-33630	84	13	in	in	ADP
bracis-33630	84	14	specific	specific	ADJ
bracis-33630	84	15	domains	domain	NOUN
bracis-33630	84	16	such	such	ADJ
bracis-33630	84	17	as	as	ADP
bracis-33630	84	18	news	news	NOUN
bracis-33630	84	19	and	and	CCONJ
bracis-33630	84	20	tweets	tweet	NOUN
bracis-33630	84	21	.	.	PUNCT
bracis-33630	85	1	this	this	DET
bracis-33630	85	2	focus	focus	NOUN
bracis-33630	85	3	has	have	AUX
bracis-33630	85	4	left	leave	VERB
bracis-33630	85	5	a	a	DET
bracis-33630	85	6	significant	significant	ADJ
bracis-33630	85	7	gap	gap	NOUN
bracis-33630	85	8	in	in	ADP
bracis-33630	85	9	our	our	PRON
bracis-33630	85	10	understanding	understanding	NOUN
bracis-33630	85	11	of	of	ADP
bracis-33630	85	12	these	these	DET
bracis-33630	85	13	methods	method	NOUN
bracis-33630	85	14	’	'	PUNCT
bracis-33630	85	15	scalability	scalability	NOUN
bracis-33630	85	16	and	and	CCONJ
bracis-33630	85	17	effectiveness	effectiveness	NOUN
bracis-33630	85	18	across	across	ADP
bracis-33630	85	19	larger	large	ADJ
bracis-33630	85	20	and	and	CCONJ
bracis-33630	85	21	more	more	ADV
bracis-33630	85	22	varied	varied	ADJ
bracis-33630	85	23	datasets	dataset	NOUN
bracis-33630	85	24	,	,	PUNCT
bracis-33630	85	25	especially	especially	ADV
bracis-33630	85	26	in	in	ADP
bracis-33630	85	27	dynamic	dynamic	ADJ
bracis-33630	85	28	sectors	sector	NOUN
bracis-33630	85	29	like	like	ADP
bracis-33630	85	30	e	e	NOUN
bracis-33630	85	31	-	-	NOUN
bracis-33630	85	32	commerce	commerce	NOUN
bracis-33630	85	33	.	.	PUNCT
bracis-33630	86	1	in	in	ADP
bracis-33630	86	2	light	light	NOUN
bracis-33630	86	3	of	of	ADP
bracis-33630	86	4	these	these	DET
bracis-33630	86	5	limitations	limitation	NOUN
bracis-33630	86	6	,	,	PUNCT
bracis-33630	86	7	our	our	PRON
bracis-33630	86	8	study	study	NOUN
bracis-33630	86	9	aims	aim	VERB
bracis-33630	86	10	to	to	PART
bracis-33630	86	11	:	:	PUNCT
bracis-33630	86	12	1	1	X
bracis-33630	86	13	.	.	X
bracis-33630	86	14	assess	assess	VERB
bracis-33630	86	15	the	the	DET
bracis-33630	86	16	method	method	NOUN
bracis-33630	86	17	’s	’s	PART
bracis-33630	86	18	effectiveness	effectiveness	NOUN
bracis-33630	86	19	on	on	ADP
bracis-33630	86	20	an	an	DET
bracis-33630	86	21	e	e	NOUN
bracis-33630	86	22	-	-	NOUN
bracis-33630	86	23	commerce	commerce	NOUN
bracis-33630	86	24	dataset	dataset	NOUN
bracis-33630	86	25	that	that	PRON
bracis-33630	86	26	varies	vary	VERB
bracis-33630	86	27	significantly	significantly	ADV
bracis-33630	86	28	in	in	ADP
bracis-33630	86	29	size	size	NOUN
bracis-33630	86	30	and	and	CCONJ
bracis-33630	86	31	number	number	NOUN
bracis-33630	86	32	of	of	ADP
bracis-33630	86	33	clusters	cluster	NOUN
bracis-33630	86	34	to	to	PART
bracis-33630	86	35	determine	determine	VERB
bracis-33630	86	36	adaptability	adaptability	NOUN
bracis-33630	86	37	and	and	CCONJ
bracis-33630	86	38	scalability	scalability	NOUN
bracis-33630	86	39	.	.	PUNCT
bracis-33630	87	1	this	this	DET
bracis-33630	87	2	evaluation	evaluation	NOUN
bracis-33630	87	3	will	will	AUX
bracis-33630	87	4	help	help	VERB
bracis-33630	87	5	us	we	PRON
bracis-33630	87	6	understand	understand	VERB
bracis-33630	87	7	how	how	SCONJ
bracis-33630	87	8	well	well	ADV
bracis-33630	87	9	the	the	DET
bracis-33630	87	10	method	method	NOUN
bracis-33630	87	11	performs	perform	VERB
bracis-33630	87	12	when	when	SCONJ
bracis-33630	87	13	subjected	subject	VERB
bracis-33630	87	14	to	to	ADP
bracis-33630	87	15	the	the	DET
bracis-33630	87	16	complexities	complexity	NOUN
bracis-33630	87	17	and	and	CCONJ
bracis-33630	87	18	varied	varied	ADJ
bracis-33630	87	19	dynamics	dynamic	NOUN
bracis-33630	87	20	of	of	ADP
bracis-33630	87	21	e	e	NOUN
bracis-33630	87	22	-	-	NOUN
bracis-33630	87	23	commerce	commerce	NOUN
bracis-33630	87	24	texts	text	NOUN
bracis-33630	87	25	compared	compare	VERB
bracis-33630	87	26	to	to	ADP
bracis-33630	87	27	traditional	traditional	ADJ
bracis-33630	87	28	short	short	ADJ
bracis-33630	87	29	text	text	NOUN
bracis-33630	87	30	streams	stream	NOUN
bracis-33630	87	31	.	.	PUNCT
bracis-33630	88	1	2	2	X
bracis-33630	88	2	.	.	X
bracis-33630	88	3	investigate	investigate	VERB
bracis-33630	88	4	the	the	DET
bracis-33630	88	5	influence	influence	NOUN
bracis-33630	88	6	of	of	ADP
bracis-33630	88	7	high	high	ADJ
bracis-33630	88	8	homogeneity	homogeneity	NOUN
bracis-33630	88	9	scores	score	NOUN
bracis-33630	88	10	on	on	ADP
bracis-33630	88	11	the	the	DET
bracis-33630	88	12	reported	report	VERB
bracis-33630	88	13	nmi	nmi	PROPN
bracis-33630	88	14	values	value	NOUN
bracis-33630	88	15	,	,	PUNCT
bracis-33630	88	16	mainly	mainly	ADV
bracis-33630	88	17	focusing	focus	VERB
bracis-33630	88	18	on	on	ADP
bracis-33630	88	19	whether	whether	SCONJ
bracis-33630	88	20	the	the	DET
bracis-33630	88	21	prevalence	prevalence	NOUN
bracis-33630	88	22	of	of	ADP
bracis-33630	88	23	single	single	ADJ
bracis-33630	88	24	-	-	PUNCT
bracis-33630	88	25	element	element	NOUN
bracis-33630	88	26	clusters	cluster	NOUN
bracis-33630	88	27	inflates	inflate	VERB
bracis-33630	88	28	these	these	DET
bracis-33630	88	29	scores	score	NOUN
bracis-33630	88	30	.	.	PUNCT
bracis-33630	89	1	this	this	DET
bracis-33630	89	2	aspect	aspect	NOUN
bracis-33630	89	3	is	be	AUX
bracis-33630	89	4	crucial	crucial	ADJ
bracis-33630	89	5	for	for	ADP
bracis-33630	89	6	ensuring	ensure	VERB
bracis-33630	89	7	that	that	SCONJ
bracis-33630	89	8	the	the	DET
bracis-33630	89	9	nmi	nmi	PROPN
bracis-33630	89	10	values	value	NOUN
bracis-33630	89	11	reflect	reflect	VERB
bracis-33630	89	12	clustering	clustering	NOUN
bracis-33630	89	13	effectiveness	effectiveness	NOUN
bracis-33630	89	14	rather	rather	ADV
bracis-33630	89	15	than	than	ADP
bracis-33630	89	16	artefacts	artefact	NOUN
bracis-33630	89	17	of	of	ADP
bracis-33630	89	18	the	the	DET
bracis-33630	89	19	clustering	clustering	ADJ
bracis-33630	89	20	process	process	NOUN
bracis-33630	89	21	.	.	PUNCT
bracis-33630	90	1	3	3	X
bracis-33630	90	2	.	.	X
bracis-33630	90	3	propose	propose	VERB
bracis-33630	90	4	and	and	CCONJ
bracis-33630	90	5	test	test	NOUN
bracis-33630	90	6	using	use	VERB
bracis-33630	90	7	the	the	DET
bracis-33630	90	8	akaike	akaike	ADJ
bracis-33630	90	9	information	information	NOUN
bracis-33630	90	10	criterion	criterion	NOUN
bracis-33630	90	11	(	(	PUNCT
bracis-33630	90	12	aic	aic	PROPN
bracis-33630	90	13	)	)	PUNCT
bracis-33630	90	14	as	as	ADP
bracis-33630	90	15	an	an	DET
bracis-33630	90	16	alternative	alternative	NOUN
bracis-33630	90	17	metric	metric	ADJ
bracis-33630	90	18	to	to	PART
bracis-33630	90	19	mitigate	mitigate	VERB
bracis-33630	90	20	the	the	DET
bracis-33630	90	21	formation	formation	NOUN
bracis-33630	90	22	of	of	ADP
bracis-33630	90	23	single	single	ADJ
bracis-33630	90	24	-	-	PUNCT
bracis-33630	90	25	element	element	NOUN
bracis-33630	90	26	clusters	cluster	NOUN
bracis-33630	90	27	.	.	PUNCT
bracis-33630	91	1	including	include	VERB
bracis-33630	91	2	aic	aic	PROPN
bracis-33630	91	3	is	be	AUX
bracis-33630	91	4	expected	expect	VERB
bracis-33630	91	5	to	to	PART
bracis-33630	91	6	provide	provide	VERB
bracis-33630	91	7	a	a	DET
bracis-33630	91	8	more	more	ADV
bracis-33630	91	9	robust	robust	ADJ
bracis-33630	91	10	evaluation	evaluation	NOUN
bracis-33630	91	11	framework	framework	NOUN
bracis-33630	91	12	that	that	PRON
bracis-33630	91	13	discourages	discourage	VERB
bracis-33630	91	14	overly	overly	ADV
bracis-33630	91	15	simplistic	simplistic	ADJ
bracis-33630	91	16	clustering	cluster	VERB
bracis-33630	91	17	solutions	solution	NOUN
bracis-33630	91	18	and	and	CCONJ
bracis-33630	91	19	promotes	promote	VERB
bracis-33630	91	20	more	more	ADV
bracis-33630	91	21	meaningful	meaningful	ADJ
bracis-33630	91	22	and	and	CCONJ
bracis-33630	91	23	practical	practical	ADJ
bracis-33630	91	24	clustering	clustering	ADJ
bracis-33630	91	25	outcomes	outcome	NOUN
bracis-33630	91	26	.	.	PUNCT
bracis-33630	92	1	the	the	DET
bracis-33630	92	2	methodology	methodology	NOUN
bracis-33630	92	3	involves	involve	VERB
bracis-33630	92	4	a	a	DET
bracis-33630	92	5	systematic	systematic	ADJ
bracis-33630	92	6	grid	grid	NOUN
bracis-33630	92	7	search	search	NOUN
bracis-33630	92	8	across	across	ADP
bracis-33630	92	9	datasets	dataset	NOUN
bracis-33630	92	10	ranging	range	VERB
bracis-33630	92	11	from	from	ADP
bracis-33630	92	12	10,000	10,000	NUM
bracis-33630	92	13	to	to	PART
bracis-33630	92	14	150,000	150,000	NUM
bracis-33630	92	15	instances	instance	NOUN
bracis-33630	92	16	to	to	PART
bracis-33630	92	17	assess	assess	VERB
bracis-33630	92	18	how	how	SCONJ
bracis-33630	92	19	varying	vary	VERB
bracis-33630	92	20	parameters	parameter	NOUN
bracis-33630	92	21	impact	impact	NOUN
bracis-33630	92	22	performance	performance	NOUN
bracis-33630	92	23	.	.	PUNCT
bracis-33630	93	1	it	it	PRON
bracis-33630	93	2	will	will	AUX
bracis-33630	93	3	lay	lay	VERB
bracis-33630	93	4	the	the	DET
bracis-33630	93	5	groundwork	groundwork	NOUN
bracis-33630	93	6	for	for	ADP
bracis-33630	93	7	a	a	DET
bracis-33630	93	8	detailed	detailed	ADJ
bracis-33630	93	9	analysis	analysis	NOUN
bracis-33630	93	10	,	,	PUNCT
bracis-33630	93	11	where	where	SCONJ
bracis-33630	93	12	each	each	DET
bracis-33630	93	13	dataset	dataset	NOUN
bracis-33630	93	14	size	size	NOUN
bracis-33630	93	15	informs	inform	VERB
bracis-33630	93	16	the	the	DET
bracis-33630	93	17	parameter	parameter	NOUN
bracis-33630	93	18	settings	setting	NOUN
bracis-33630	93	19	for	for	ADP
bracis-33630	93	20	the	the	DET
bracis-33630	93	21	next	next	ADJ
bracis-33630	93	22	,	,	PUNCT
bracis-33630	93	23	ensuring	ensure	VERB
bracis-33630	93	24	continuous	continuous	ADJ
bracis-33630	93	25	refinement	refinement	NOUN
bracis-33630	93	26	and	and	CCONJ
bracis-33630	93	27	optimization	optimization	NOUN
bracis-33630	93	28	of	of	ADP
bracis-33630	93	29	both	both	CCONJ
bracis-33630	93	30	aic	aic	PROPN
bracis-33630	93	31	and	and	CCONJ
bracis-33630	93	32	nmi	nmi	PROPN
bracis-33630	93	33	.	.	PUNCT
bracis-33630	94	1	the	the	DET
bracis-33630	94	2	aic	aic	PROPN
bracis-33630	94	3	optimization	optimization	NOUN
bracis-33630	94	4	aims	aim	VERB
bracis-33630	94	5	to	to	PART
bracis-33630	94	6	minimize	minimize	VERB
bracis-33630	94	7	the	the	DET
bracis-33630	94	8	aic	aic	PROPN
bracis-33630	94	9	value	value	NOUN
bracis-33630	94	10	,	,	PUNCT
bracis-33630	94	11	while	while	SCONJ
bracis-33630	94	12	the	the	DET
bracis-33630	94	13	nmi	nmi	PROPN
bracis-33630	94	14	optimization	optimization	NOUN
bracis-33630	94	15	aims	aim	VERB
bracis-33630	94	16	to	to	PART
bracis-33630	94	17	maximize	maximize	VERB
bracis-33630	94	18	the	the	DET
bracis-33630	94	19	nmi	nmi	PROPN
bracis-33630	94	20	score	score	NOUN
bracis-33630	94	21	.	.	PUNCT
bracis-33630	95	1	this	this	DET
bracis-33630	95	2	approach	approach	NOUN
bracis-33630	95	3	seeks	seek	VERB
bracis-33630	95	4	to	to	PART
bracis-33630	95	5	provide	provide	VERB
bracis-33630	95	6	insights	insight	NOUN
bracis-33630	95	7	into	into	ADP
bracis-33630	95	8	the	the	DET
bracis-33630	95	9	method	method	NOUN
bracis-33630	95	10	’s	’s	PART
bracis-33630	95	11	performance	performance	NOUN
bracis-33630	95	12	across	across	ADP
bracis-33630	95	13	various	various	ADJ
bracis-33630	95	14	scenarios	scenario	NOUN
bracis-33630	95	15	and	and	CCONJ
bracis-33630	95	16	improve	improve	VERB
bracis-33630	95	17	clustering	clustering	NOUN
bracis-33630	95	18	practices	practice	NOUN
bracis-33630	95	19	.	.	PUNCT
bracis-33630	96	1	3.1	3.1	NUM
bracis-33630	96	2	datasets	dataset	NOUN
bracis-33630	96	3	our	our	PRON
bracis-33630	96	4	experimental	experimental	ADJ
bracis-33630	96	5	study	study	NOUN
bracis-33630	96	6	employed	employ	VERB
bracis-33630	96	7	a	a	DET
bracis-33630	96	8	set	set	NOUN
bracis-33630	96	9	of	of	ADP
bracis-33630	96	10	datasets	dataset	NOUN
bracis-33630	96	11	based	base	VERB
bracis-33630	96	12	on	on	ADP
bracis-33630	96	13	the	the	DET
bracis-33630	96	14	brazilian	brazilian	ADJ
bracis-33630	96	15	nf	nf	NOUN
bracis-33630	96	16	-	-	PUNCT
bracis-33630	96	17	e	e	NOUN
bracis-33630	96	18	project	project	NOUN
bracis-33630	96	19	,	,	PUNCT
bracis-33630	96	20	initiated	initiate	VERB
bracis-33630	96	21	in	in	ADP
bracis-33630	96	22	2006	2006	NUM
bracis-33630	96	23	,	,	PUNCT
bracis-33630	96	24	which	which	PRON
bracis-33630	96	25	revolutionized	revolutionize	VERB
bracis-33630	96	26	tax	tax	NOUN
bracis-33630	96	27	documentation	documentation	NOUN
bracis-33630	96	28	by	by	ADP
bracis-33630	96	29	transitioning	transition	VERB
bracis-33630	96	30	to	to	ADP
bracis-33630	96	31	an	an	DET
bracis-33630	96	32	electronic	electronic	ADJ
bracis-33630	96	33	system	system	NOUN
bracis-33630	96	34	.	.	PUNCT
bracis-33630	97	1	this	this	DET
bracis-33630	97	2	shift	shift	NOUN
bracis-33630	97	3	has	have	AUX
bracis-33630	97	4	generated	generate	VERB
bracis-33630	97	5	a	a	DET
bracis-33630	97	6	large	large	ADJ
bracis-33630	97	7	volume	volume	NOUN
bracis-33630	97	8	of	of	ADP
bracis-33630	97	9	data	datum	NOUN
bracis-33630	97	10	,	,	PUNCT
bracis-33630	97	11	including	include	VERB
bracis-33630	97	12	comprehensive	comprehensive	ADJ
bracis-33630	97	13	invoice	invoice	NOUN
bracis-33630	97	14	details	detail	NOUN
bracis-33630	97	15	,	,	PUNCT
bracis-33630	97	16	making	make	VERB
bracis-33630	97	17	it	it	PRON
bracis-33630	97	18	a	a	DET
bracis-33630	97	19	valuable	valuable	ADJ
bracis-33630	97	20	resource	resource	NOUN
bracis-33630	97	21	for	for	ADP
bracis-33630	97	22	machine	machine	NOUN
bracis-33630	97	23	learning	learn	VERB
bracis-33630	97	24	applications	application	NOUN
bracis-33630	97	25	,	,	PUNCT
bracis-33630	97	26	such	such	ADJ
bracis-33630	97	27	as	as	ADP
bracis-33630	97	28	detecting	detect	VERB
bracis-33630	97	29	tax	tax	NOUN
bracis-33630	97	30	fraud	fraud	NOUN
bracis-33630	97	31	.	.	PUNCT
bracis-33630	98	1	we	we	PRON
bracis-33630	98	2	analyzed	analyze	VERB
bracis-33630	98	3	a	a	DET
bracis-33630	98	4	subset	subset	NOUN
bracis-33630	98	5	of	of	ADP
bracis-33630	98	6	the	the	DET
bracis-33630	98	7	nf	nf	NOUN
bracis-33630	98	8	-	-	PUNCT
bracis-33630	98	9	e	e	NOUN
bracis-33630	98	10	dataset	dataset	NOUN
bracis-33630	98	11	from	from	ADP
bracis-33630	98	12	november	november	PROPN
bracis-33630	98	13	2021	2021	NUM
bracis-33630	98	14	,	,	PUNCT
bracis-33630	98	15	provided	provide	VERB
bracis-33630	98	16	by	by	ADP
bracis-33630	98	17	the	the	DET
bracis-33630	98	18	amazonas	amazonas	PROPN
bracis-33630	98	19	state	state	PROPN
bracis-33630	98	20	department	department	PROPN
bracis-33630	98	21	of	of	ADP
bracis-33630	98	22	finance	finance	NOUN
bracis-33630	98	23	.	.	PUNCT
bracis-33630	99	1	the	the	DET
bracis-33630	99	2	dataset	dataset	NOUN
bracis-33630	99	3	contains	contain	VERB
bracis-33630	99	4	several	several	ADJ
bracis-33630	99	5	key	key	ADJ
bracis-33630	99	6	features	feature	NOUN
bracis-33630	99	7	,	,	PUNCT
bracis-33630	99	8	including	include	VERB
bracis-33630	99	9	:	:	PUNCT
bracis-33630	99	10	gtin	gtin	NOUN
bracis-33630	99	11	:	:	PUNCT
bracis-33630	99	12	the	the	DET
bracis-33630	99	13	gtin	gtin	NOUN
bracis-33630	99	14	is	be	AUX
bracis-33630	99	15	an	an	DET
bracis-33630	99	16	identifier	identifier	NOUN
bracis-33630	99	17	for	for	ADP
bracis-33630	99	18	trade	trade	NOUN
bracis-33630	99	19	items	item	NOUN
bracis-33630	99	20	developed	develop	VERB
bracis-33630	99	21	and	and	CCONJ
bracis-33630	99	22	controlled	control	VERB
bracis-33630	99	23	by	by	ADP
bracis-33630	99	24	gs1	gs1	PROPN
bracis-33630	99	25	,	,	PUNCT
bracis-33630	99	26	formerly	formerly	ADV
bracis-33630	99	27	ean	ean	NOUN
bracis-33630	99	28	/	/	SYM
bracis-33630	99	29	ucc	ucc	PROPN
bracis-33630	99	30	.	.	PROPN
bracis-33630	99	31	gtins	gtins	PROPN
bracis-33630	99	32	,	,	PUNCT
bracis-33630	99	33	formerly	formerly	ADV
bracis-33630	99	34	called	call	VERB
bracis-33630	99	35	ean	ean	NOUN
bracis-33630	99	36	codes	code	NOUN
bracis-33630	99	37	,	,	PUNCT
bracis-33630	99	38	are	be	AUX
bracis-33630	99	39	assigned	assign	VERB
bracis-33630	99	40	to	to	ADP
bracis-33630	99	41	any	any	DET
bracis-33630	99	42	item	item	NOUN
bracis-33630	99	43	(	(	PUNCT
bracis-33630	99	44	product	product	NOUN
bracis-33630	99	45	or	or	CCONJ
bracis-33630	99	46	service	service	NOUN
bracis-33630	99	47	)	)	PUNCT
bracis-33630	99	48	that	that	PRON
bracis-33630	99	49	can	can	AUX
bracis-33630	99	50	be	be	AUX
bracis-33630	99	51	priced	price	VERB
bracis-33630	99	52	,	,	PUNCT
bracis-33630	99	53	ordered	order	VERB
bracis-33630	99	54	,	,	PUNCT
bracis-33630	99	55	or	or	CCONJ
bracis-33630	99	56	invoiced	invoice	VERB
bracis-33630	99	57	at	at	ADP
bracis-33630	99	58	any	any	DET
bracis-33630	99	59	point	point	NOUN
bracis-33630	99	60	in	in	ADP
bracis-33630	99	61	the	the	DET
bracis-33630	99	62	supply	supply	NOUN
bracis-33630	99	63	chain	chain	NOUN
bracis-33630	99	64	.	.	PUNCT
bracis-33630	100	1	ncm	ncm	PROPN
bracis-33630	100	2	:	:	PUNCT
bracis-33630	100	3	the	the	DET
bracis-33630	100	4	ncmfootnote	ncmfootnote	NOUN
bracis-33630	100	5	1	1	NUM
bracis-33630	100	6	,	,	PUNCT
bracis-33630	100	7	which	which	PRON
bracis-33630	100	8	is	be	AUX
bracis-33630	100	9	a	a	DET
bracis-33630	100	10	regional	regional	ADJ
bracis-33630	100	11	nomenclature	nomenclature	NOUN
bracis-33630	100	12	for	for	ADP
bracis-33630	100	13	the	the	DET
bracis-33630	100	14	categorization	categorization	NOUN
bracis-33630	100	15	of	of	ADP
bracis-33630	100	16	goods	good	NOUN
bracis-33630	100	17	adopted	adopt	VERB
bracis-33630	100	18	by	by	ADP
bracis-33630	100	19	since	since	SCONJ
bracis-33630	100	20	1995	1995	NUM
bracis-33630	100	21	,	,	PUNCT
bracis-33630	100	22	being	be	AUX
bracis-33630	100	23	used	use	VERB
bracis-33630	100	24	in	in	ADP
bracis-33630	100	25	all	all	DET
bracis-33630	100	26	foreign	foreign	ADJ
bracis-33630	100	27	trade	trade	NOUN
bracis-33630	100	28	operations	operation	NOUN
bracis-33630	100	29	of	of	ADP
bracis-33630	100	30	mercosur	mercosur	PROPN
bracis-33630	100	31	countries	country	NOUN
bracis-33630	100	32	.	.	PUNCT
bracis-33630	101	1	the	the	DET
bracis-33630	101	2	ncm	ncm	PROPN
bracis-33630	101	3	is	be	AUX
bracis-33630	101	4	based	base	VERB
bracis-33630	101	5	on	on	ADP
bracis-33630	101	6	the	the	DET
bracis-33630	101	7	hs	hs	PROPN
bracis-33630	101	8	,	,	PUNCT
bracis-33630	101	9	a	a	DET
bracis-33630	101	10	condensed	condense	VERB
bracis-33630	101	11	expression	expression	NOUN
bracis-33630	101	12	of	of	ADP
bracis-33630	101	13	the	the	DET
bracis-33630	101	14	“	"	PUNCT
bracis-33630	101	15	harmonized	harmonized	ADJ
bracis-33630	101	16	commodity	commodity	NOUN
bracis-33630	101	17	description	description	NOUN
bracis-33630	101	18	and	and	CCONJ
bracis-33630	101	19	coding	code	VERB
bracis-33630	101	20	system	system	NOUN
bracis-33630	101	21	”	"	PUNCT
bracis-33630	101	22	maintained	maintain	VERB
bracis-33630	101	23	by	by	ADP
bracis-33630	101	24	the	the	DET
bracis-33630	101	25	wco	wco	NOUN
bracis-33630	101	26	.	.	PUNCT
bracis-33630	102	1	it	it	PRON
bracis-33630	102	2	was	be	AUX
bracis-33630	102	3	created	create	VERB
bracis-33630	102	4	to	to	PART
bracis-33630	102	5	improve	improve	VERB
bracis-33630	102	6	and	and	CCONJ
bracis-33630	102	7	facilitate	facilitate	VERB
bracis-33630	102	8	international	international	ADJ
bracis-33630	102	9	trade	trade	NOUN
bracis-33630	102	10	and	and	CCONJ
bracis-33630	102	11	its	its	PRON
bracis-33630	102	12	statistical	statistical	ADJ
bracis-33630	102	13	control	control	NOUN
bracis-33630	102	14	.	.	PUNCT
bracis-33630	103	1	product	product	NOUN
bracis-33630	103	2	description	description	NOUN
bracis-33630	103	3	:	:	PUNCT
bracis-33630	103	4	a	a	DET
bracis-33630	103	5	free	free	ADJ
bracis-33630	103	6	text	text	NOUN
bracis-33630	103	7	field	field	NOUN
bracis-33630	103	8	on	on	ADP
bracis-33630	103	9	the	the	DET
bracis-33630	103	10	invoice	invoice	NOUN
bracis-33630	103	11	with	with	ADP
bracis-33630	103	12	a	a	DET
bracis-33630	103	13	brief	brief	ADJ
bracis-33630	103	14	product	product	NOUN
bracis-33630	103	15	description	description	NOUN
bracis-33630	103	16	.	.	PUNCT
bracis-33630	104	1	table	table	NOUN
bracis-33630	104	2	1	1	NUM
bracis-33630	104	3	.	.	PUNCT
bracis-33630	105	1	examples	example	NOUN
bracis-33630	105	2	of	of	ADP
bracis-33630	105	3	brazilian	brazilian	ADJ
bracis-33630	105	4	products	product	NOUN
bracis-33630	105	5	with	with	ADP
bracis-33630	105	6	gtin	gtin	NOUN
bracis-33630	105	7	and	and	CCONJ
bracis-33630	105	8	ncm	ncm	PROPN
bracis-33630	105	9	.	.	PUNCT
bracis-33630	105	10	ncm	ncm	PROPN
bracis-33630	105	11	codes	code	NOUN
bracis-33630	105	12	22030000	22030000	NUM
bracis-33630	105	13	(	(	PUNCT
bracis-33630	105	14	general	general	ADJ
bracis-33630	105	15	beer	beer	NOUN
bracis-33630	105	16	packaging	packaging	NOUN
bracis-33630	105	17	)	)	PUNCT
bracis-33630	105	18	and	and	CCONJ
bracis-33630	105	19	22030010	22030010	NUM
bracis-33630	105	20	(	(	PUNCT
bracis-33630	105	21	specific	specific	ADJ
bracis-33630	105	22	types	type	NOUN
bracis-33630	105	23	like	like	ADP
bracis-33630	105	24	wheat	wheat	NOUN
bracis-33630	105	25	beer	beer	NOUN
bracis-33630	105	26	)	)	PUNCT
bracis-33630	105	27	.	.	PUNCT
bracis-33630	106	1	both	both	PRON
bracis-33630	106	2	share	share	VERB
bracis-33630	106	3	ncm4	ncm4	PROPN
bracis-33630	106	4	2203	2203	NUM
bracis-33630	106	5	(	(	PUNCT
bracis-33630	106	6	“	"	PUNCT
bracis-33630	106	7	beer	beer	NOUN
bracis-33630	106	8	”	"	PUNCT
bracis-33630	106	9	)	)	PUNCT
bracis-33630	106	10	,	,	PUNCT
bracis-33630	106	11	but	but	CCONJ
bracis-33630	106	12	the	the	DET
bracis-33630	106	13	additional	additional	ADJ
bracis-33630	106	14	digits	digit	NOUN
bracis-33630	106	15	provide	provide	VERB
bracis-33630	106	16	specific	specific	ADJ
bracis-33630	106	17	details	detail	NOUN
bracis-33630	106	18	about	about	ADP
bracis-33630	106	19	type	type	NOUN
bracis-33630	106	20	and	and	CCONJ
bracis-33630	106	21	packagingfull	packagingfull	ADJ
bracis-33630	106	22	size	size	NOUN
bracis-33630	106	23	table	table	NOUN
bracis-33630	106	24	3.2	3.2	NUM
bracis-33630	106	25	data	datum	NOUN
bracis-33630	106	26	pre	pre	ADJ
bracis-33630	106	27	-	-	NOUN
bracis-33630	106	28	processing	processing	ADJ
bracis-33630	106	29	to	to	PART
bracis-33630	106	30	better	well	ADV
bracis-33630	106	31	understand	understand	VERB
bracis-33630	106	32	the	the	DET
bracis-33630	106	33	intrinsic	intrinsic	ADJ
bracis-33630	106	34	properties	property	NOUN
bracis-33630	106	35	of	of	ADP
bracis-33630	106	36	our	our	PRON
bracis-33630	106	37	dataset	dataset	NOUN
bracis-33630	106	38	,	,	PUNCT
bracis-33630	106	39	we	we	PRON
bracis-33630	106	40	have	have	AUX
bracis-33630	106	41	segmented	segment	VERB
bracis-33630	106	42	the	the	DET
bracis-33630	106	43	data	datum	NOUN
bracis-33630	106	44	into	into	ADP
bracis-33630	106	45	manageable	manageable	ADJ
bracis-33630	106	46	blocks	block	NOUN
bracis-33630	106	47	,	,	PUNCT
bracis-33630	106	48	increasing	increase	VERB
bracis-33630	106	49	in	in	ADP
bracis-33630	106	50	increments	increment	NOUN
bracis-33630	106	51	of	of	ADP
bracis-33630	106	52	10,000	10,000	NUM
bracis-33630	106	53	up	up	ADP
bracis-33630	106	54	to	to	ADP
bracis-33630	106	55	a	a	DET
bracis-33630	106	56	maximum	maximum	NOUN
bracis-33630	106	57	of	of	ADP
bracis-33630	106	58	150,000	150,000	NUM
bracis-33630	106	59	instances	instance	NOUN
bracis-33630	106	60	.	.	PUNCT
bracis-33630	107	1	this	this	DET
bracis-33630	107	2	segmentation	segmentation	NOUN
bracis-33630	107	3	is	be	AUX
bracis-33630	107	4	designed	design	VERB
bracis-33630	107	5	to	to	PART
bracis-33630	107	6	facilitate	facilitate	VERB
bracis-33630	107	7	analysis	analysis	NOUN
bracis-33630	107	8	and	and	CCONJ
bracis-33630	107	9	ensure	ensure	VERB
bracis-33630	107	10	each	each	DET
bracis-33630	107	11	subset	subset	NOUN
bracis-33630	107	12	remains	remain	VERB
bracis-33630	107	13	a	a	DET
bracis-33630	107	14	practical	practical	ADJ
bracis-33630	107	15	size	size	NOUN
bracis-33630	107	16	for	for	ADP
bracis-33630	107	17	detailed	detailed	ADJ
bracis-33630	107	18	cluster	cluster	NOUN
bracis-33630	107	19	evaluation	evaluation	NOUN
bracis-33630	107	20	.	.	PUNCT
bracis-33630	108	1	in	in	ADP
bracis-33630	108	2	our	our	PRON
bracis-33630	108	3	analysis	analysis	NOUN
bracis-33630	108	4	,	,	PUNCT
bracis-33630	108	5	we	we	PRON
bracis-33630	108	6	distinguish	distinguish	VERB
bracis-33630	108	7	between	between	ADP
bracis-33630	108	8	two	two	NUM
bracis-33630	108	9	coding	code	VERB
bracis-33630	108	10	formats	format	NOUN
bracis-33630	108	11	used	use	VERB
bracis-33630	108	12	for	for	ADP
bracis-33630	108	13	dataset	dataset	ADJ
bracis-33630	108	14	generation	generation	NOUN
bracis-33630	108	15	:	:	PUNCT
bracis-33630	108	16	ncm4	ncm4	PROPN
bracis-33630	108	17	and	and	CCONJ
bracis-33630	108	18	ncm8	ncm8	PROPN
bracis-33630	108	19	.	.	PUNCT
bracis-33630	109	1	the	the	DET
bracis-33630	109	2	ncm4	ncm4	PROPN
bracis-33630	109	3	code	code	NOUN
bracis-33630	109	4	signifies	signify	VERB
bracis-33630	109	5	broader	broad	ADJ
bracis-33630	109	6	product	product	NOUN
bracis-33630	109	7	categories	category	NOUN
bracis-33630	109	8	using	use	VERB
bracis-33630	109	9	just	just	ADV
bracis-33630	109	10	the	the	DET
bracis-33630	109	11	first	first	ADJ
bracis-33630	109	12	four	four	NUM
bracis-33630	109	13	digits	digit	NOUN
bracis-33630	109	14	,	,	PUNCT
bracis-33630	109	15	providing	provide	VERB
bracis-33630	109	16	a	a	DET
bracis-33630	109	17	general	general	ADJ
bracis-33630	109	18	view	view	NOUN
bracis-33630	109	19	of	of	ADP
bracis-33630	109	20	item	item	NOUN
bracis-33630	109	21	groups	group	NOUN
bracis-33630	109	22	.	.	PUNCT
bracis-33630	110	1	in	in	ADP
bracis-33630	110	2	contrast	contrast	NOUN
bracis-33630	110	3	,	,	PUNCT
bracis-33630	110	4	the	the	DET
bracis-33630	110	5	ncm8	ncm8	PROPN
bracis-33630	110	6	code	code	PROPN
bracis-33630	110	7	utilizes	utilize	VERB
bracis-33630	110	8	all	all	DET
bracis-33630	110	9	eight	eight	NUM
bracis-33630	110	10	digits	digit	NOUN
bracis-33630	110	11	for	for	ADP
bracis-33630	110	12	a	a	DET
bracis-33630	110	13	more	more	ADV
bracis-33630	110	14	detailed	detailed	ADJ
bracis-33630	110	15	and	and	CCONJ
bracis-33630	110	16	specific	specific	ADJ
bracis-33630	110	17	classification	classification	NOUN
bracis-33630	110	18	,	,	PUNCT
bracis-33630	110	19	revealing	reveal	VERB
bracis-33630	110	20	finer	fine	ADJ
bracis-33630	110	21	distinctions	distinction	NOUN
bracis-33630	110	22	in	in	ADP
bracis-33630	110	23	product	product	NOUN
bracis-33630	110	24	types	type	NOUN
bracis-33630	110	25	.	.	PUNCT
bracis-33630	111	1	fig	fig	NOUN
bracis-33630	111	2	.	.	PUNCT
bracis-33630	112	1	1	1	X
bracis-33630	112	2	.	.	X
bracis-33630	112	3	characteristics	characteristic	NOUN
bracis-33630	112	4	of	of	ADP
bracis-33630	112	5	the	the	DET
bracis-33630	112	6	datasets	dataset	NOUN
bracis-33630	112	7	:	:	PUNCT
bracis-33630	112	8	number	number	NOUN
bracis-33630	112	9	of	of	ADP
bracis-33630	112	10	documents	document	NOUN
bracis-33630	112	11	(	(	PUNCT
bracis-33630	112	12	d	d	NOUN
bracis-33630	112	13	)	)	PUNCT
bracis-33630	112	14	,	,	PUNCT
bracis-33630	112	15	number	number	NOUN
bracis-33630	112	16	of	of	ADP
bracis-33630	112	17	clusters	cluster	NOUN
bracis-33630	112	18	(	(	PUNCT
bracis-33630	112	19	k	k	NOUN
bracis-33630	112	20	)	)	PUNCT
bracis-33630	112	21	,	,	PUNCT
bracis-33630	112	22	vocabulary	vocabulary	ADJ
bracis-33630	112	23	size	size	NOUN
bracis-33630	112	24	(	(	PUNCT
bracis-33630	112	25	v	v	NOUN
bracis-33630	112	26	)	)	PUNCT
bracis-33630	112	27	,	,	PUNCT
bracis-33630	112	28	average	average	ADJ
bracis-33630	112	29	length	length	NOUN
bracis-33630	112	30	of	of	ADP
bracis-33630	112	31	the	the	DET
bracis-33630	112	32	documents	document	NOUN
bracis-33630	112	33	(	(	PUNCT
bracis-33630	112	34	avglen	avglen	PROPN
bracis-33630	112	35	)	)	PUNCT
bracis-33630	112	36	.	.	PUNCT
bracis-33630	113	1	full	full	ADJ
bracis-33630	113	2	size	size	NOUN
bracis-33630	113	3	image	image	NOUN
bracis-33630	113	4	the	the	DET
bracis-33630	113	5	behavior	behavior	NOUN
bracis-33630	113	6	of	of	ADP
bracis-33630	113	7	the	the	DET
bracis-33630	113	8	metric	metric	ADJ
bracis-33630	113	9	‘	'	PUNCT
bracis-33630	113	10	k	k	NOUN
bracis-33630	113	11	’	'	PUNCT
bracis-33630	113	12	,	,	PUNCT
bracis-33630	113	13	which	which	PRON
bracis-33630	113	14	represents	represent	VERB
bracis-33630	113	15	the	the	DET
bracis-33630	113	16	number	number	NOUN
bracis-33630	113	17	of	of	ADP
bracis-33630	113	18	clusters	cluster	NOUN
bracis-33630	113	19	,	,	PUNCT
bracis-33630	113	20	varies	vary	VERB
bracis-33630	113	21	significantly	significantly	ADV
bracis-33630	113	22	across	across	ADP
bracis-33630	113	23	different	different	ADJ
bracis-33630	113	24	data	datum	NOUN
bracis-33630	113	25	groupings	grouping	NOUN
bracis-33630	113	26	.	.	PUNCT
bracis-33630	114	1	for	for	ADP
bracis-33630	114	2	ncm4	ncm4	PROPN
bracis-33630	114	3	,	,	PUNCT
bracis-33630	114	4	there	there	PRON
bracis-33630	114	5	is	be	VERB
bracis-33630	114	6	a	a	DET
bracis-33630	114	7	gradual	gradual	ADJ
bracis-33630	114	8	increase	increase	NOUN
bracis-33630	114	9	in	in	ADP
bracis-33630	114	10	‘	'	PUNCT
bracis-33630	114	11	k	k	NOUN
bracis-33630	114	12	’	'	PUNCT
bracis-33630	114	13	,	,	PUNCT
bracis-33630	114	14	marked	mark	VERB
bracis-33630	114	15	by	by	ADP
bracis-33630	114	16	sporadic	sporadic	ADJ
bracis-33630	114	17	larger	large	ADJ
bracis-33630	114	18	increases	increase	NOUN
bracis-33630	114	19	that	that	DET
bracis-33630	114	20	hint	hint	NOUN
bracis-33630	114	21	at	at	ADP
bracis-33630	114	22	emerging	emerge	VERB
bracis-33630	114	23	complexities	complexity	NOUN
bracis-33630	114	24	within	within	ADP
bracis-33630	114	25	the	the	DET
bracis-33630	114	26	data	datum	NOUN
bracis-33630	114	27	as	as	SCONJ
bracis-33630	114	28	the	the	DET
bracis-33630	114	29	volume	volume	NOUN
bracis-33630	114	30	grows	grow	VERB
bracis-33630	114	31	.	.	PUNCT
bracis-33630	115	1	the	the	DET
bracis-33630	115	2	ncm8	ncm8	PROPN
bracis-33630	115	3	displays	display	VERB
bracis-33630	115	4	a	a	DET
bracis-33630	115	5	more	more	ADV
bracis-33630	115	6	uniform	uniform	ADJ
bracis-33630	115	7	rise	rise	NOUN
bracis-33630	115	8	in	in	ADP
bracis-33630	115	9	‘	'	PUNCT
bracis-33630	115	10	k	k	NOUN
bracis-33630	115	11	’	'	PUNCT
bracis-33630	115	12	,	,	PUNCT
bracis-33630	115	13	suggesting	suggest	VERB
bracis-33630	115	14	a	a	DET
bracis-33630	115	15	linear	linear	ADJ
bracis-33630	115	16	enhancement	enhancement	NOUN
bracis-33630	115	17	in	in	ADP
bracis-33630	115	18	topic	topic	NOUN
bracis-33630	115	19	diversity	diversity	NOUN
bracis-33630	115	20	corresponding	correspond	VERB
bracis-33630	115	21	to	to	ADP
bracis-33630	115	22	data	datum	NOUN
bracis-33630	115	23	size	size	NOUN
bracis-33630	115	24	.	.	PUNCT
bracis-33630	116	1	tailored	tailor	VERB
bracis-33630	116	2	to	to	ADP
bracis-33630	116	3	specific	specific	ADJ
bracis-33630	116	4	product	product	NOUN
bracis-33630	116	5	descriptions	description	NOUN
bracis-33630	116	6	,	,	PUNCT
bracis-33630	116	7	gtin	gtin	NOUN
bracis-33630	116	8	clusters	cluster	NOUN
bracis-33630	116	9	show	show	VERB
bracis-33630	116	10	a	a	DET
bracis-33630	116	11	steep	steep	ADJ
bracis-33630	116	12	upward	upward	ADJ
bracis-33630	116	13	trend	trend	NOUN
bracis-33630	116	14	in	in	ADP
bracis-33630	116	15	‘	'	PUNCT
bracis-33630	116	16	k	k	NOUN
bracis-33630	116	17	’	'	PUNCT
bracis-33630	116	18	.	.	PUNCT
bracis-33630	117	1	this	this	PRON
bracis-33630	117	2	indicates	indicate	VERB
bracis-33630	117	3	a	a	DET
bracis-33630	117	4	widening	widen	VERB
bracis-33630	117	5	range	range	NOUN
bracis-33630	117	6	of	of	ADP
bracis-33630	117	7	topics	topic	NOUN
bracis-33630	117	8	and	and	CCONJ
bracis-33630	117	9	reflects	reflect	VERB
bracis-33630	117	10	gtin	gtin	NOUN
bracis-33630	117	11	’s	’s	PART
bracis-33630	117	12	detailed	detailed	ADJ
bracis-33630	117	13	and	and	CCONJ
bracis-33630	117	14	expansive	expansive	ADJ
bracis-33630	117	15	nature	nature	NOUN
bracis-33630	117	16	,	,	PUNCT
bracis-33630	117	17	where	where	SCONJ
bracis-33630	117	18	adding	add	VERB
bracis-33630	117	19	more	more	ADJ
bracis-33630	117	20	product	product	NOUN
bracis-33630	117	21	specifics	specific	NOUN
bracis-33630	117	22	leads	lead	VERB
bracis-33630	117	23	to	to	ADP
bracis-33630	117	24	greater	great	ADJ
bracis-33630	117	25	diversity	diversity	NOUN
bracis-33630	117	26	in	in	ADP
bracis-33630	117	27	the	the	DET
bracis-33630	117	28	identified	identify	VERB
bracis-33630	117	29	topics	topic	NOUN
bracis-33630	117	30	.	.	PUNCT
bracis-33630	118	1	regarding	regard	VERB
bracis-33630	118	2	the	the	DET
bracis-33630	118	3	vocabulary	vocabulary	ADJ
bracis-33630	118	4	size	size	NOUN
bracis-33630	118	5	‘	'	PUNCT
bracis-33630	118	6	v	v	NOUN
bracis-33630	118	7	’	'	PUNCT
bracis-33630	118	8	,	,	PUNCT
bracis-33630	118	9	the	the	DET
bracis-33630	118	10	gtin	gtin	NOUN
bracis-33630	118	11	dataset	dataset	VERB
bracis-33630	118	12	shows	show	VERB
bracis-33630	118	13	a	a	DET
bracis-33630	118	14	pronounced	pronounced	ADJ
bracis-33630	118	15	growth	growth	NOUN
bracis-33630	118	16	,	,	PUNCT
bracis-33630	118	17	indicating	indicate	VERB
bracis-33630	118	18	an	an	DET
bracis-33630	118	19	expanding	expand	VERB
bracis-33630	118	20	lexicon	lexicon	NOUN
bracis-33630	118	21	that	that	PRON
bracis-33630	118	22	accommodates	accommodate	VERB
bracis-33630	118	23	the	the	DET
bracis-33630	118	24	increasing	increase	VERB
bracis-33630	118	25	diversity	diversity	NOUN
bracis-33630	118	26	of	of	ADP
bracis-33630	118	27	products	product	NOUN
bracis-33630	118	28	.	.	PUNCT
bracis-33630	119	1	however	however	ADV
bracis-33630	119	2	,	,	PUNCT
bracis-33630	119	3	for	for	ADP
bracis-33630	119	4	ncm	ncm	PROPN
bracis-33630	119	5	subsets	subset	NOUN
bracis-33630	119	6	—	—	PUNCT
bracis-33630	119	7	ncm4	ncm4	PROPN
bracis-33630	119	8	and	and	CCONJ
bracis-33630	119	9	ncm8	ncm8	PROPN
bracis-33630	119	10	—	—	PUNCT
bracis-33630	119	11	the	the	DET
bracis-33630	119	12	growth	growth	NOUN
bracis-33630	119	13	in	in	ADP
bracis-33630	119	14	‘	'	PUNCT
bracis-33630	119	15	v	v	NOUN
bracis-33630	119	16	’	'	PUNCT
bracis-33630	119	17	is	be	AUX
bracis-33630	119	18	more	more	ADV
bracis-33630	119	19	measured	measure	VERB
bracis-33630	119	20	,	,	PUNCT
bracis-33630	119	21	underscoring	underscore	VERB
bracis-33630	119	22	that	that	SCONJ
bracis-33630	119	23	while	while	SCONJ
bracis-33630	119	24	gtin	gtin	NOUN
bracis-33630	119	25	descriptions	description	NOUN
bracis-33630	119	26	might	might	AUX
bracis-33630	119	27	capture	capture	VERB
bracis-33630	119	28	a	a	DET
bracis-33630	119	29	larger	large	ADJ
bracis-33630	119	30	number	number	NOUN
bracis-33630	119	31	of	of	ADP
bracis-33630	119	32	clusters	cluster	NOUN
bracis-33630	119	33	due	due	ADP
bracis-33630	119	34	to	to	ADP
bracis-33630	119	35	unique	unique	ADJ
bracis-33630	119	36	product	product	NOUN
bracis-33630	119	37	variations	variation	NOUN
bracis-33630	119	38	,	,	PUNCT
bracis-33630	119	39	a	a	DET
bracis-33630	119	40	single	single	ADJ
bracis-33630	119	41	ncm	ncm	PROPN
bracis-33630	119	42	code	code	NOUN
bracis-33630	119	43	can	can	AUX
bracis-33630	119	44	encompass	encompass	VERB
bracis-33630	119	45	a	a	DET
bracis-33630	119	46	broad	broad	ADJ
bracis-33630	119	47	range	range	NOUN
bracis-33630	119	48	of	of	ADP
bracis-33630	119	49	products	product	NOUN
bracis-33630	119	50	,	,	PUNCT
bracis-33630	119	51	hence	hence	ADV
bracis-33630	119	52	a	a	DET
bracis-33630	119	53	larger	large	ADJ
bracis-33630	119	54	vocabulary	vocabulary	NOUN
bracis-33630	119	55	.	.	PUNCT
bracis-33630	120	1	finally	finally	ADV
bracis-33630	120	2	,	,	PUNCT
bracis-33630	120	3	in	in	ADP
bracis-33630	120	4	our	our	PRON
bracis-33630	120	5	data	data	NOUN
bracis-33630	120	6	preparation	preparation	NOUN
bracis-33630	120	7	,	,	PUNCT
bracis-33630	120	8	we	we	PRON
bracis-33630	120	9	carefully	carefully	ADV
bracis-33630	120	10	control	control	VERB
bracis-33630	120	11	the	the	DET
bracis-33630	120	12	average	average	ADJ
bracis-33630	120	13	document	document	NOUN
bracis-33630	120	14	length	length	NOUN
bracis-33630	120	15	(	(	PUNCT
bracis-33630	120	16	‘	'	PUNCT
bracis-33630	120	17	avg	avg	NOUN
bracis-33630	120	18	len	len	ADJ
bracis-33630	120	19	’	'	PUNCT
bracis-33630	120	20	)	)	PUNCT
bracis-33630	120	21	to	to	PART
bracis-33630	120	22	ensure	ensure	VERB
bracis-33630	120	23	that	that	SCONJ
bracis-33630	120	24	clusters	cluster	NOUN
bracis-33630	120	25	are	be	AUX
bracis-33630	120	26	not	not	PART
bracis-33630	120	27	too	too	ADV
bracis-33630	120	28	small	small	ADJ
bracis-33630	120	29	or	or	CCONJ
bracis-33630	120	30	too	too	ADV
bracis-33630	120	31	large	large	ADJ
bracis-33630	120	32	,	,	PUNCT
bracis-33630	120	33	which	which	PRON
bracis-33630	120	34	could	could	AUX
bracis-33630	120	35	skew	skew	VERB
bracis-33630	120	36	the	the	DET
bracis-33630	120	37	analysis	analysis	NOUN
bracis-33630	120	38	.	.	PUNCT
bracis-33630	121	1	the	the	DET
bracis-33630	121	2	‘	'	PUNCT
bracis-33630	121	3	avg	avg	NOUN
bracis-33630	121	4	len	len	ADJ
bracis-33630	121	5	’	'	PUNCT
bracis-33630	121	6	metric	metric	NOUN
bracis-33630	121	7	shows	show	VERB
bracis-33630	121	8	minor	minor	ADJ
bracis-33630	121	9	variations	variation	NOUN
bracis-33630	121	10	across	across	ADP
bracis-33630	121	11	gtin	gtin	NOUN
bracis-33630	121	12	and	and	CCONJ
bracis-33630	121	13	ncm	ncm	PROPN
bracis-33630	121	14	datasets	dataset	NOUN
bracis-33630	121	15	,	,	PUNCT
bracis-33630	121	16	with	with	ADP
bracis-33630	121	17	a	a	DET
bracis-33630	121	18	slight	slight	ADJ
bracis-33630	121	19	decrease	decrease	NOUN
bracis-33630	121	20	for	for	ADP
bracis-33630	121	21	gtin	gtin	NOUN
bracis-33630	121	22	,	,	PUNCT
bracis-33630	121	23	suggesting	suggest	VERB
bracis-33630	121	24	that	that	SCONJ
bracis-33630	121	25	product	product	NOUN
bracis-33630	121	26	descriptions	description	NOUN
bracis-33630	121	27	become	become	VERB
bracis-33630	121	28	more	more	ADV
bracis-33630	121	29	concise	concise	ADJ
bracis-33630	121	30	as	as	SCONJ
bracis-33630	121	31	the	the	DET
bracis-33630	121	32	dataset	dataset	NOUN
bracis-33630	121	33	grows	grow	VERB
bracis-33630	121	34	.	.	PUNCT
bracis-33630	122	1	however	however	ADV
bracis-33630	122	2	,	,	PUNCT
bracis-33630	122	3	for	for	ADP
bracis-33630	122	4	ncm4	ncm4	NOUN
bracis-33630	122	5	and	and	CCONJ
bracis-33630	122	6	ncm8	ncm8	PROPN
bracis-33630	122	7	,	,	PUNCT
bracis-33630	122	8	there	there	PRON
bracis-33630	122	9	are	be	VERB
bracis-33630	122	10	only	only	ADV
bracis-33630	122	11	slight	slight	ADJ
bracis-33630	122	12	fluctuations	fluctuation	NOUN
bracis-33630	122	13	,	,	PUNCT
bracis-33630	122	14	reflecting	reflect	VERB
bracis-33630	122	15	the	the	DET
bracis-33630	122	16	diverse	diverse	ADJ
bracis-33630	122	17	complexity	complexity	NOUN
bracis-33630	122	18	within	within	ADP
bracis-33630	122	19	their	their	PRON
bracis-33630	122	20	broader	broad	ADJ
bracis-33630	122	21	item	item	NOUN
bracis-33630	122	22	categories	category	NOUN
bracis-33630	122	23	as	as	SCONJ
bracis-33630	122	24	they	they	PRON
bracis-33630	122	25	expand	expand	VERB
bracis-33630	122	26	.	.	PUNCT
bracis-33630	123	1	this	this	DET
bracis-33630	123	2	control	control	NOUN
bracis-33630	123	3	of	of	ADP
bracis-33630	123	4	‘	'	PUNCT
bracis-33630	123	5	avg	avg	NOUN
bracis-33630	123	6	len	len	ADJ
bracis-33630	123	7	’	'	PUNCT
bracis-33630	123	8	is	be	AUX
bracis-33630	123	9	crucial	crucial	ADJ
bracis-33630	123	10	for	for	ADP
bracis-33630	123	11	maintaining	maintain	VERB
bracis-33630	123	12	the	the	DET
bracis-33630	123	13	integrity	integrity	NOUN
bracis-33630	123	14	and	and	CCONJ
bracis-33630	123	15	comparability	comparability	NOUN
bracis-33630	123	16	of	of	ADP
bracis-33630	123	17	clusters	cluster	NOUN
bracis-33630	123	18	across	across	ADP
bracis-33630	123	19	different	different	ADJ
bracis-33630	123	20	dataset	dataset	NOUN
bracis-33630	123	21	sizes	size	NOUN
bracis-33630	123	22	.	.	PUNCT
bracis-33630	124	1	3.3	3.3	NUM
bracis-33630	124	2	clustering	clustering	NOUN
bracis-33630	124	3	methods	method	NOUN
bracis-33630	124	4	online	online	VERB
bracis-33630	124	5	semantic	semantic	ADV
bracis-33630	124	6	-	-	PUNCT
bracis-33630	124	7	enhanced	enhance	VERB
bracis-33630	124	8	dirichlet	dirichlet	PROPN
bracis-33630	124	9	model	model	NOUN
bracis-33630	124	10	(	(	PUNCT
bracis-33630	124	11	osdm	osdm	PROPN
bracis-33630	124	12	)	)	PUNCT
bracis-33630	125	1	[	[	X
bracis-33630	125	2	19	19	NUM
bracis-33630	125	3	]	]	PUNCT
bracis-33630	125	4	:	:	PUNCT
bracis-33630	125	5	features	feature	VERB
bracis-33630	125	6	a	a	DET
bracis-33630	125	7	dynamic	dynamic	ADJ
bracis-33630	125	8	clustering	clustering	NOUN
bracis-33630	125	9	approach	approach	NOUN
bracis-33630	125	10	,	,	PUNCT
bracis-33630	125	11	integrating	integrate	VERB
bracis-33630	125	12	word	word	NOUN
bracis-33630	125	13	-	-	PUNCT
bracis-33630	125	14	occurrence	occurrence	NOUN
bracis-33630	125	15	semantic	semantic	ADJ
bracis-33630	125	16	information	information	NOUN
bracis-33630	125	17	into	into	ADP
bracis-33630	125	18	a	a	DET
bracis-33630	125	19	graphical	graphical	ADJ
bracis-33630	125	20	model	model	NOUN
bracis-33630	125	21	.	.	PUNCT
bracis-33630	126	1	it	it	PRON
bracis-33630	126	2	allows	allow	VERB
bracis-33630	126	3	for	for	ADP
bracis-33630	126	4	real	real	ADJ
bracis-33630	126	5	-	-	PUNCT
bracis-33630	126	6	time	time	NOUN
bracis-33630	126	7	text	text	NOUN
bracis-33630	126	8	processing	processing	NOUN
bracis-33630	126	9	without	without	ADP
bracis-33630	126	10	the	the	DET
bracis-33630	126	11	need	need	NOUN
bracis-33630	126	12	for	for	ADP
bracis-33630	126	13	predetermined	predetermine	VERB
bracis-33630	126	14	cluster	cluster	NOUN
bracis-33630	126	15	numbers	number	NOUN
bracis-33630	126	16	or	or	CCONJ
bracis-33630	126	17	batch	batch	NOUN
bracis-33630	126	18	sizes	size	NOUN
bracis-33630	126	19	,	,	PUNCT
bracis-33630	126	20	showcasing	showcase	VERB
bracis-33630	126	21	its	its	PRON
bracis-33630	126	22	adaptability	adaptability	NOUN
bracis-33630	126	23	to	to	ADP
bracis-33630	126	24	varying	vary	VERB
bracis-33630	126	25	data	datum	NOUN
bracis-33630	126	26	streams.footnote	streams.footnote	VERB
bracis-33630	126	27	2	2	NUM
bracis-33630	126	28	online	online	ADJ
bracis-33630	126	29	semantic	semantic	ADV
bracis-33630	126	30	-	-	PUNCT
bracis-33630	126	31	enhanced	enhance	VERB
bracis-33630	126	32	graphical	graphical	ADJ
bracis-33630	126	33	model	model	NOUN
bracis-33630	126	34	(	(	PUNCT
bracis-33630	126	35	osgm	osgm	NOUN
bracis-33630	126	36	)	)	PUNCT
bracis-33630	127	1	[	[	X
bracis-33630	127	2	20	20	NUM
bracis-33630	127	3	]	]	PUNCT
bracis-33630	127	4	:	:	PUNCT
bracis-33630	127	5	uses	use	VERB
bracis-33630	127	6	a	a	DET
bracis-33630	127	7	poly	poly	ADJ
bracis-33630	127	8	urn	urn	NOUN
bracis-33630	127	9	scheme	scheme	NOUN
bracis-33630	127	10	for	for	ADP
bracis-33630	127	11	dynamic	dynamic	ADJ
bracis-33630	127	12	document	document	NOUN
bracis-33630	127	13	clustering	clustering	NOUN
bracis-33630	127	14	,	,	PUNCT
bracis-33630	127	15	focusing	focus	VERB
bracis-33630	127	16	on	on	ADP
bracis-33630	127	17	semantic	semantic	ADJ
bracis-33630	127	18	smoothing	smoothing	NOUN
bracis-33630	127	19	and	and	CCONJ
bracis-33630	127	20	term	term	NOUN
bracis-33630	127	21	co	co	NOUN
bracis-33630	127	22	-	-	NOUN
bracis-33630	127	23	occurrence	occurrence	NOUN
bracis-33630	127	24	to	to	PART
bracis-33630	127	25	resolve	resolve	VERB
bracis-33630	127	26	term	term	NOUN
bracis-33630	127	27	ambiguity	ambiguity	NOUN
bracis-33630	127	28	and	and	CCONJ
bracis-33630	127	29	enhance	enhance	VERB
bracis-33630	127	30	clarity	clarity	NOUN
bracis-33630	127	31	,	,	PUNCT
bracis-33630	127	32	proving	prove	VERB
bracis-33630	127	33	its	its	PRON
bracis-33630	127	34	strength	strength	NOUN
bracis-33630	127	35	in	in	ADP
bracis-33630	127	36	semantic	semantic	NOUN
bracis-33630	127	37	-	-	PUNCT
bracis-33630	127	38	based	base	VERB
bracis-33630	127	39	text	text	NOUN
bracis-33630	127	40	clustering.footnote	clustering.footnote	VERB
bracis-33630	127	41	3	3	NUM
bracis-33630	127	42	eindm	eindm	NOUN
bracis-33630	128	1	[	[	X
bracis-33630	128	2	21	21	NUM
bracis-33630	128	3	]	]	X
bracis-33630	128	4	:	:	PUNCT
bracis-33630	128	5	a	a	DET
bracis-33630	128	6	dirichlet	dirichlet	PROPN
bracis-33630	128	7	model	model	NOUN
bracis-33630	128	8	using	use	VERB
bracis-33630	128	9	window	window	NOUN
bracis-33630	128	10	-	-	PUNCT
bracis-33630	128	11	based	base	VERB
bracis-33630	128	12	semantic	semantic	ADJ
bracis-33630	128	13	term	term	NOUN
bracis-33630	128	14	representation	representation	NOUN
bracis-33630	128	15	to	to	PART
bracis-33630	128	16	capture	capture	VERB
bracis-33630	128	17	the	the	DET
bracis-33630	128	18	evolving	evolve	VERB
bracis-33630	128	19	nature	nature	NOUN
bracis-33630	128	20	of	of	ADP
bracis-33630	128	21	text	text	NOUN
bracis-33630	128	22	data	datum	NOUN
bracis-33630	128	23	through	through	ADP
bracis-33630	128	24	enhanced	enhance	VERB
bracis-33630	128	25	semantic	semantic	ADJ
bracis-33630	128	26	understanding	understanding	NOUN
bracis-33630	128	27	and	and	CCONJ
bracis-33630	128	28	novel	novel	NOUN
bracis-33630	128	29	clustering	clustering	NOUN
bracis-33630	128	30	metrics.footnote	metrics.footnote	NOUN
bracis-33630	128	31	4	4	NUM
bracis-33630	128	32	estream	estream	NOUN
bracis-33630	128	33	[	[	X
bracis-33630	128	34	22	22	NUM
bracis-33630	128	35	]	]	PUNCT
bracis-33630	128	36	:	:	PUNCT
bracis-33630	128	37	employs	employ	VERB
bracis-33630	128	38	both	both	CCONJ
bracis-33630	128	39	online	online	ADJ
bracis-33630	128	40	and	and	CCONJ
bracis-33630	128	41	offline	offline	ADJ
bracis-33630	128	42	clustering	cluster	VERB
bracis-33630	128	43	to	to	PART
bracis-33630	128	44	effectively	effectively	ADV
bracis-33630	128	45	adapt	adapt	VERB
bracis-33630	128	46	to	to	ADP
bracis-33630	128	47	various	various	ADJ
bracis-33630	128	48	text	text	NOUN
bracis-33630	128	49	streaming	streaming	NOUN
bracis-33630	128	50	scenarios.footnote	scenarios.footnote	ADJ
bracis-33630	128	51	5	5	NUM
bracis-33630	128	52	3.4	3.4	NUM
bracis-33630	128	53	evaluation	evaluation	NOUN
bracis-33630	128	54	metrics	metric	NOUN
bracis-33630	128	55	to	to	PART
bracis-33630	128	56	evaluate	evaluate	VERB
bracis-33630	128	57	our	our	PRON
bracis-33630	128	58	experimental	experimental	ADJ
bracis-33630	128	59	results	result	NOUN
bracis-33630	128	60	,	,	PUNCT
bracis-33630	128	61	we	we	PRON
bracis-33630	128	62	resorted	resort	VERB
bracis-33630	128	63	 	 	SPACE
bracis-33630	128	64	to	to	ADP
bracis-33630	128	65	metrics	metric	NOUN
bracis-33630	128	66	widely	widely	ADV
bracis-33630	128	67	used	use	VERB
bracis-33630	128	68	in	in	ADP
bracis-33630	128	69	the	the	DET
bracis-33630	128	70	literature	literature	NOUN
bracis-33630	128	71	[	[	X
bracis-33630	128	72	23	23	NUM
bracis-33630	128	73	]	]	PUNCT
bracis-33630	128	74	to	to	PART
bracis-33630	128	75	provide	provide	VERB
bracis-33630	128	76	a	a	DET
bracis-33630	128	77	comprehensive	comprehensive	ADJ
bracis-33630	128	78	assessment	assessment	NOUN
bracis-33630	128	79	of	of	ADP
bracis-33630	128	80	clustering	clustering	ADJ
bracis-33630	128	81	performance	performance	NOUN
bracis-33630	128	82	and	and	CCONJ
bracis-33630	128	83	model	model	NOUN
bracis-33630	128	84	evaluation	evaluation	NOUN
bracis-33630	128	85	.	.	PUNCT
bracis-33630	129	1	the	the	DET
bracis-33630	129	2	metrics	metric	NOUN
bracis-33630	129	3	are	be	AUX
bracis-33630	129	4	the	the	DET
bracis-33630	129	5	following	follow	VERB
bracis-33630	129	6	:	:	PUNCT
bracis-33630	129	7	homogeneity	homogeneity	NOUN
bracis-33630	129	8	(	(	PUNCT
bracis-33630	129	9	h	h	NOUN
bracis-33630	129	10	)	)	PUNCT
bracis-33630	129	11	measures	measure	NOUN
bracis-33630	129	12	how	how	SCONJ
bracis-33630	129	13	well	well	ADV
bracis-33630	129	14	each	each	DET
bracis-33630	129	15	cluster	cluster	NOUN
bracis-33630	129	16	contains	contain	VERB
bracis-33630	129	17	only	only	ADV
bracis-33630	129	18	members	member	NOUN
bracis-33630	129	19	of	of	ADP
bracis-33630	129	20	a	a	DET
bracis-33630	129	21	single	single	ADJ
bracis-33630	129	22	class	class	NOUN
bracis-33630	129	23	.	.	PUNCT
bracis-33630	130	1	it	it	PRON
bracis-33630	130	2	is	be	AUX
bracis-33630	130	3	defined	define	VERB
bracis-33630	130	4	as	as	ADP
bracis-33630	130	5	:	:	PUNCT
bracis-33630	130	6	$	$	SYM
bracis-33630	130	7	$	$	SYM
bracis-33630	130	8	\begin{aligned	\begin{aligne	VERB
bracis-33630	130	9	}	}	PUNCT
bracis-33630	130	10	h	h	NOUN
bracis-33630	130	11	=	=	SYM
bracis-33630	130	12	1	1	NUM
bracis-33630	130	13	\frac{h(u|v)}{h(u	\frac{h(u|v)}{h(u	NOUN
bracis-33630	130	14	)	)	PUNCT
bracis-33630	130	15	}	}	PUNCT
bracis-33630	130	16	\end{aligned}$$	\end{aligned}$$	X
bracis-33630	130	17	(	(	PUNCT
bracis-33630	130	18	1	1	NUM
bracis-33630	130	19	)	)	PUNCT
bracis-33630	130	20	where	where	SCONJ
bracis-33630	130	21	\	\	PROPN
bracis-33630	130	22	(	(	PUNCT
bracis-33630	130	23	h(u|v	h(u|v	X
bracis-33630	130	24	)	)	PUNCT
bracis-33630	130	25	\	\	NOUN
bracis-33630	130	26	)	)	PUNCT
bracis-33630	130	27	is	be	AUX
bracis-33630	130	28	the	the	DET
bracis-33630	130	29	conditional	conditional	ADJ
bracis-33630	130	30	entropy	entropy	NOUN
bracis-33630	130	31	of	of	ADP
bracis-33630	130	32	the	the	DET
bracis-33630	130	33	class	class	NOUN
bracis-33630	130	34	distribution	distribution	NOUN
bracis-33630	130	35	given	give	VERB
bracis-33630	130	36	the	the	DET
bracis-33630	130	37	cluster	cluster	NOUN
bracis-33630	130	38	assignments	assignment	NOUN
bracis-33630	130	39	,	,	PUNCT
bracis-33630	130	40	and	and	CCONJ
bracis-33630	130	41	\	\	PROPN
bracis-33630	130	42	(	(	PUNCT
bracis-33630	130	43	h(u	h(u	PROPN
bracis-33630	130	44	)	)	PUNCT
bracis-33630	130	45	\	\	NOUN
bracis-33630	130	46	)	)	PUNCT
bracis-33630	130	47	is	be	AUX
bracis-33630	130	48	the	the	DET
bracis-33630	130	49	entropy	entropy	NOUN
bracis-33630	130	50	of	of	ADP
bracis-33630	130	51	the	the	DET
bracis-33630	130	52	class	class	NOUN
bracis-33630	130	53	distribution	distribution	NOUN
bracis-33630	130	54	.	.	PUNCT
bracis-33630	131	1	completeness	completeness	NOUN
bracis-33630	131	2	(	(	PUNCT
bracis-33630	131	3	c	c	NOUN
bracis-33630	131	4	)	)	PUNCT
bracis-33630	131	5	evaluates	evaluate	VERB
bracis-33630	131	6	how	how	SCONJ
bracis-33630	131	7	well	well	ADV
bracis-33630	131	8	each	each	DET
bracis-33630	131	9	class	class	NOUN
bracis-33630	131	10	is	be	AUX
bracis-33630	131	11	represented	represent	VERB
bracis-33630	131	12	within	within	ADP
bracis-33630	131	13	a	a	DET
bracis-33630	131	14	single	single	ADJ
bracis-33630	131	15	cluster	cluster	NOUN
bracis-33630	131	16	.	.	PUNCT
bracis-33630	132	1	it	it	PRON
bracis-33630	132	2	is	be	AUX
bracis-33630	132	3	defined	define	VERB
bracis-33630	132	4	as	as	ADP
bracis-33630	132	5	:	:	PUNCT
bracis-33630	132	6	$	$	SYM
bracis-33630	132	7	$	$	SYM
bracis-33630	132	8	\begin{aligned	\begin{aligne	VERB
bracis-33630	132	9	}	}	PUNCT
bracis-33630	132	10	c	c	NOUN
bracis-33630	132	11	=	=	SYM
bracis-33630	132	12	1	1	NUM
bracis-33630	132	13	\frac{h(v|u)}{h(v	\frac{h(v|u)}{h(v	NOUN
bracis-33630	132	14	)	)	PUNCT
bracis-33630	132	15	}	}	PUNCT
bracis-33630	132	16	\end{aligned}$$	\end{aligned}$$	X
bracis-33630	132	17	(	(	PUNCT
bracis-33630	132	18	2	2	NUM
bracis-33630	132	19	)	)	PUNCT
bracis-33630	132	20	where	where	SCONJ
bracis-33630	132	21	\	\	PROPN
bracis-33630	132	22	(	(	PUNCT
bracis-33630	132	23	h(v|u	h(v|u	NUM
bracis-33630	132	24	)	)	PUNCT
bracis-33630	132	25	\	\	NOUN
bracis-33630	132	26	)	)	PUNCT
bracis-33630	132	27	is	be	AUX
bracis-33630	132	28	the	the	DET
bracis-33630	132	29	conditional	conditional	ADJ
bracis-33630	132	30	entropy	entropy	NOUN
bracis-33630	132	31	of	of	ADP
bracis-33630	132	32	the	the	DET
bracis-33630	132	33	cluster	cluster	NOUN
bracis-33630	132	34	distribution	distribution	NOUN
bracis-33630	132	35	given	give	VERB
bracis-33630	132	36	the	the	DET
bracis-33630	132	37	class	class	NOUN
bracis-33630	132	38	assignments	assignment	NOUN
bracis-33630	132	39	,	,	PUNCT
bracis-33630	132	40	and	and	CCONJ
bracis-33630	132	41	\	\	PROPN
bracis-33630	132	42	(	(	PUNCT
bracis-33630	132	43	h(v	h(v	PROPN
bracis-33630	132	44	)	)	PUNCT
bracis-33630	132	45	\	\	NOUN
bracis-33630	132	46	)	)	PUNCT
bracis-33630	132	47	is	be	AUX
bracis-33630	132	48	the	the	DET
bracis-33630	132	49	entropy	entropy	NOUN
bracis-33630	132	50	of	of	ADP
bracis-33630	132	51	the	the	DET
bracis-33630	132	52	cluster	cluster	NOUN
bracis-33630	132	53	distribution	distribution	NOUN
bracis-33630	132	54	.	.	PUNCT
bracis-33630	133	1	normalized	normalize	VERB
bracis-33630	133	2	mutual	mutual	ADJ
bracis-33630	133	3	information	information	NOUN
bracis-33630	133	4	(	(	PUNCT
bracis-33630	133	5	nmi	nmi	PROPN
bracis-33630	133	6	)	)	PUNCT
bracis-33630	133	7	combines	combine	VERB
bracis-33630	133	8	homogeneity	homogeneity	NOUN
bracis-33630	133	9	and	and	CCONJ
bracis-33630	133	10	completeness	completeness	NOUN
bracis-33630	133	11	,	,	PUNCT
bracis-33630	133	12	measuring	measure	VERB
bracis-33630	133	13	the	the	DET
bracis-33630	133	14	similarity	similarity	NOUN
bracis-33630	133	15	between	between	ADP
bracis-33630	133	16	the	the	DET
bracis-33630	133	17	true	true	ADJ
bracis-33630	133	18	class	class	NOUN
bracis-33630	133	19	labels	label	NOUN
bracis-33630	133	20	and	and	CCONJ
bracis-33630	133	21	the	the	DET
bracis-33630	133	22	assigned	assign	VERB
bracis-33630	133	23	cluster	cluster	NOUN
bracis-33630	133	24	labels	label	NOUN
bracis-33630	133	25	.	.	PUNCT
bracis-33630	134	1	it	it	PRON
bracis-33630	134	2	is	be	AUX
bracis-33630	134	3	defined	define	VERB
bracis-33630	134	4	as	as	ADP
bracis-33630	134	5	:	:	PUNCT
bracis-33630	134	6	$	$	SYM
bracis-33630	134	7	$	$	SYM
bracis-33630	134	8	\begin{aligned	\begin{aligne	VERB
bracis-33630	134	9	}	}	PUNCT
bracis-33630	134	10	nmi	nmi	NOUN
bracis-33630	134	11	=	=	SYM
bracis-33630	134	12	\frac{2	\frac{2	PROPN
bracis-33630	134	13	\cdot	\cdot	NOUN
bracis-33630	134	14	i(u;v)}{h(u	i(u;v)}{h(u	PROPN
bracis-33630	134	15	)	)	PUNCT
bracis-33630	135	1	+	+	CCONJ
bracis-33630	135	2	h(v	h(v	NOUN
bracis-33630	135	3	)	)	PUNCT
bracis-33630	135	4	}	}	PUNCT
bracis-33630	135	5	\end{aligned}$$	\end{aligned}$$	X
bracis-33630	135	6	(	(	PUNCT
bracis-33630	135	7	3	3	NUM
bracis-33630	135	8	)	)	PUNCT
bracis-33630	135	9	where	where	SCONJ
bracis-33630	135	10	\	\	PROPN
bracis-33630	135	11	(	(	PUNCT
bracis-33630	135	12	i(u;v	i(u;v	ADJ
bracis-33630	135	13	)	)	PUNCT
bracis-33630	135	14	\	\	NOUN
bracis-33630	135	15	)	)	PUNCT
bracis-33630	135	16	is	be	AUX
bracis-33630	135	17	the	the	DET
bracis-33630	135	18	mutual	mutual	ADJ
bracis-33630	135	19	information	information	NOUN
bracis-33630	135	20	between	between	ADP
bracis-33630	135	21	class	class	NOUN
bracis-33630	135	22	and	and	CCONJ
bracis-33630	135	23	cluster	cluster	NOUN
bracis-33630	135	24	distribution	distribution	NOUN
bracis-33630	135	25	.	.	PUNCT
bracis-33630	136	1	purity	purity	NOUN
bracis-33630	136	2	(	(	PUNCT
bracis-33630	136	3	p	p	NOUN
bracis-33630	136	4	)	)	PUNCT
bracis-33630	136	5	assesses	assess	VERB
bracis-33630	136	6	the	the	DET
bracis-33630	136	7	clustering	cluster	VERB
bracis-33630	136	8	quality	quality	NOUN
bracis-33630	136	9	by	by	ADP
bracis-33630	136	10	calculating	calculate	VERB
bracis-33630	136	11	the	the	DET
bracis-33630	136	12	proportion	proportion	NOUN
bracis-33630	136	13	of	of	ADP
bracis-33630	136	14	the	the	DET
bracis-33630	136	15	dominant	dominant	ADJ
bracis-33630	136	16	class	class	NOUN
bracis-33630	136	17	in	in	ADP
bracis-33630	136	18	each	each	DET
bracis-33630	136	19	cluster	cluster	NOUN
bracis-33630	136	20	.	.	PUNCT
bracis-33630	137	1	it	it	PRON
bracis-33630	137	2	is	be	AUX
bracis-33630	137	3	defined	define	VERB
bracis-33630	137	4	as	as	ADP
bracis-33630	137	5	:	:	PUNCT
bracis-33630	137	6	$	$	SYM
bracis-33630	137	7	$	$	SYM
bracis-33630	137	8	\begin{aligned	\begin{aligne	VERB
bracis-33630	137	9	}	}	PUNCT
bracis-33630	137	10	p	p	NOUN
bracis-33630	137	11	=	=	PROPN
bracis-33630	137	12	\frac{1}{n	\frac{1}{n	NOUN
bracis-33630	137	13	}	}	PUNCT
bracis-33630	137	14	\sum	\sum	NOUN
bracis-33630	138	1	_	_	PUNCT
bracis-33630	138	2	{	{	PUNCT
bracis-33630	138	3	k	k	X
bracis-33630	138	4	}	}	PUNCT
bracis-33630	138	5	\max	\max	PROPN
bracis-33630	138	6	_	_	PUNCT
bracis-33630	138	7	{	{	PUNCT
bracis-33630	138	8	j	j	NOUN
bracis-33630	138	9	}	}	PUNCT
bracis-33630	138	10	|	|	NOUN
bracis-33630	138	11	c_k	c_k	PUNCT
bracis-33630	138	12	\cap	\cap	PROPN
bracis-33630	138	13	t_j	t_j	PUNCT
bracis-33630	139	1	|	|	ADV
bracis-33630	139	2	\end{aligned}$$	\end{aligned}$$	X
bracis-33630	139	3	(	(	PUNCT
bracis-33630	139	4	4	4	NUM
bracis-33630	139	5	)	)	PUNCT
bracis-33630	139	6	where	where	SCONJ
bracis-33630	139	7	\	\	PROPN
bracis-33630	139	8	(	(	PUNCT
bracis-33630	139	9	n	n	PRON
bracis-33630	139	10	\	\	NOUN
bracis-33630	139	11	)	)	PUNCT
bracis-33630	139	12	is	be	AUX
bracis-33630	139	13	the	the	DET
bracis-33630	139	14	total	total	ADJ
bracis-33630	139	15	number	number	NOUN
bracis-33630	139	16	of	of	ADP
bracis-33630	139	17	samples	sample	NOUN
bracis-33630	139	18	,	,	PUNCT
bracis-33630	139	19	\	\	PROPN
bracis-33630	139	20	(	(	PUNCT
bracis-33630	139	21	c_k	c_k	SYM
bracis-33630	139	22	\	\	NOUN
bracis-33630	139	23	)	)	PUNCT
bracis-33630	139	24	is	be	AUX
bracis-33630	139	25	the	the	DET
bracis-33630	139	26	set	set	NOUN
bracis-33630	139	27	of	of	ADP
bracis-33630	139	28	samples	sample	NOUN
bracis-33630	139	29	in	in	ADP
bracis-33630	139	30	cluster	cluster	NOUN
bracis-33630	139	31	\	\	PROPN
bracis-33630	139	32	(	(	PUNCT
bracis-33630	139	33	k	k	PROPN
bracis-33630	139	34	\	\	PROPN
bracis-33630	139	35	)	)	PUNCT
bracis-33630	139	36	,	,	PUNCT
bracis-33630	139	37	and	and	CCONJ
bracis-33630	139	38	\	\	PROPN
bracis-33630	139	39	(	(	PUNCT
bracis-33630	139	40	t_j	t_j	SYM
bracis-33630	139	41	\	\	PROPN
bracis-33630	139	42	)	)	PUNCT
bracis-33630	139	43	is	be	AUX
bracis-33630	139	44	the	the	DET
bracis-33630	139	45	set	set	NOUN
bracis-33630	139	46	of	of	ADP
bracis-33630	139	47	samples	sample	NOUN
bracis-33630	139	48	in	in	ADP
bracis-33630	139	49	class	class	NOUN
bracis-33630	139	50	\	\	PROPN
bracis-33630	139	51	(	(	PUNCT
bracis-33630	139	52	j	j	PROPN
bracis-33630	139	53	\	\	PROPN
bracis-33630	139	54	)	)	PUNCT
bracis-33630	139	55	.	.	PUNCT
bracis-33630	140	1	akaike	akaike	ADJ
bracis-33630	140	2	information	information	NOUN
bracis-33630	140	3	criterion	criterion	NOUN
bracis-33630	140	4	(	(	PUNCT
bracis-33630	140	5	aic	aic	PROPN
bracis-33630	140	6	)	)	PUNCT
bracis-33630	140	7	helps	help	VERB
bracis-33630	140	8	balance	balance	NOUN
bracis-33630	140	9	model	model	NOUN
bracis-33630	140	10	fit	fit	ADJ
bracis-33630	140	11	and	and	CCONJ
bracis-33630	140	12	complexity	complexity	NOUN
bracis-33630	140	13	.	.	PUNCT
bracis-33630	141	1	it	it	PRON
bracis-33630	141	2	is	be	AUX
bracis-33630	141	3	defined	define	VERB
bracis-33630	141	4	as	as	ADP
bracis-33630	141	5	:	:	PUNCT
bracis-33630	141	6	$	$	SYM
bracis-33630	141	7	$	$	SYM
bracis-33630	141	8	\begin{aligned	\begin{aligne	VERB
bracis-33630	141	9	}	}	PUNCT
bracis-33630	141	10	aic	aic	PROPN
bracis-33630	141	11	=	=	SYM
bracis-33630	141	12	2ptr2\ln	2ptr2\ln	NUM
bracis-33630	141	13	(	(	PUNCT
bracis-33630	141	14	l	l	NOUN
bracis-33630	141	15	)	)	PUNCT
bracis-33630	141	16	\end{aligned}$$	\end{aligned}$$	X
bracis-33630	141	17	(	(	PUNCT
bracis-33630	141	18	5	5	NUM
bracis-33630	141	19	)	)	PUNCT
bracis-33630	141	20	where	where	SCONJ
bracis-33630	141	21	\	\	PROPN
bracis-33630	141	22	(	(	PUNCT
bracis-33630	141	23	{	{	PUNCT
bracis-33630	141	24	ptr	ptr	PROPN
bracis-33630	141	25	}	}	PUNCT
bracis-33630	141	26	\	\	PROPN
bracis-33630	141	27	)	)	PUNCT
bracis-33630	141	28	represents	represent	VERB
bracis-33630	141	29	the	the	DET
bracis-33630	141	30	predict	predict	ADJ
bracis-33630	141	31	/	/	SYM
bracis-33630	141	32	true	true	ADJ
bracis-33630	141	33	size	size	NOUN
bracis-33630	141	34	rate	rate	NOUN
bracis-33630	141	35	,	,	PUNCT
bracis-33630	141	36	and	and	CCONJ
bracis-33630	141	37	\	\	PROPN
bracis-33630	141	38	(	(	PUNCT
bracis-33630	141	39	l	l	NOUN
bracis-33630	141	40	\	\	PROPN
bracis-33630	141	41	)	)	PUNCT
bracis-33630	141	42	is	be	AUX
bracis-33630	141	43	the	the	DET
bracis-33630	141	44	likelihood	likelihood	NOUN
bracis-33630	141	45	of	of	ADP
bracis-33630	141	46	observing	observe	VERB
bracis-33630	141	47	the	the	DET
bracis-33630	141	48	homogeneity	homogeneity	NOUN
bracis-33630	141	49	value	value	NOUN
bracis-33630	141	50	\	\	PROPN
bracis-33630	141	51	(	(	PUNCT
bracis-33630	141	52	h	h	NOUN
bracis-33630	141	53	\	\	PROPN
bracis-33630	141	54	)	)	PUNCT
bracis-33630	141	55	given	give	VERB
bracis-33630	141	56	the	the	DET
bracis-33630	141	57	mean	mean	NOUN
bracis-33630	141	58	(	(	PUNCT
bracis-33630	141	59	\	\	PROPN
bracis-33630	141	60	(	(	PUNCT
bracis-33630	141	61	\mu	\mu	PROPN
bracis-33630	141	62	\	\	PROPN
bracis-33630	141	63	)	)	PUNCT
bracis-33630	141	64	)	)	PUNCT
bracis-33630	141	65	and	and	CCONJ
bracis-33630	141	66	standard	standard	ADJ
bracis-33630	141	67	deviation	deviation	NOUN
bracis-33630	141	68	(	(	PUNCT
bracis-33630	141	69	\	\	PROPN
bracis-33630	141	70	(	(	PUNCT
bracis-33630	141	71	\sigma	\sigma	PROPN
bracis-33630	141	72	\	\	PROPN
bracis-33630	141	73	)	)	PUNCT
bracis-33630	141	74	)	)	PUNCT
bracis-33630	141	75	of	of	ADP
bracis-33630	141	76	the	the	DET
bracis-33630	141	77	h	h	NOUN
bracis-33630	141	78	values	value	NOUN
bracis-33630	141	79	across	across	ADP
bracis-33630	141	80	the	the	DET
bracis-33630	141	81	datasets	dataset	NOUN
bracis-33630	141	82	.	.	PUNCT
bracis-33630	142	1	the	the	DET
bracis-33630	142	2	likelihood	likelihood	NOUN
bracis-33630	142	3	is	be	AUX
bracis-33630	142	4	computed	compute	VERB
bracis-33630	142	5	using	use	VERB
bracis-33630	142	6	the	the	DET
bracis-33630	142	7	normal	normal	ADJ
bracis-33630	142	8	distribution	distribution	NOUN
bracis-33630	142	9	:	:	PUNCT
bracis-33630	142	10	$	$	SYM
bracis-33630	142	11	$	$	SYM
bracis-33630	142	12	\begin{aligned	\begin{aligne	VERB
bracis-33630	142	13	}	}	PUNCT
bracis-33630	142	14	l	l	NOUN
bracis-33630	142	15	=	=	SYM
bracis-33630	142	16	\frac{1}{\sqrt{2	\frac{1}{\sqrt{2	ADJ
bracis-33630	142	17	\pi	\pi	PROPN
bracis-33630	142	18	}	}	PUNCT
bracis-33630	142	19	\sigma	\sigma	PROPN
bracis-33630	142	20	}	}	PUNCT
bracis-33630	142	21	\exp	\exp	PROPN
bracis-33630	142	22	\left	\left	PROPN
bracis-33630	142	23	(	(	PUNCT
bracis-33630	142	24	-\frac{(h	-\frac{(h	PROPN
bracis-33630	142	25	\mu	\mu	PROPN
bracis-33630	142	26	)	)	PUNCT
bracis-33630	142	27	^2}{2	^2}{2	X
bracis-33630	142	28	\sigma	\sigma	PROPN
bracis-33630	142	29	^2	^2	NUM
bracis-33630	142	30	}	}	SYM
bracis-33630	142	31	\right	\right	PROPN
bracis-33630	142	32	)	)	PUNCT
bracis-33630	142	33	\end{aligned}$$	\end{aligned}$$	X
bracis-33630	142	34	(	(	PUNCT
bracis-33630	142	35	6	6	NUM
bracis-33630	142	36	)	)	PUNCT
bracis-33630	142	37	the	the	DET
bracis-33630	142	38	predict	predict	ADJ
bracis-33630	142	39	/	/	SYM
bracis-33630	142	40	true	true	ADJ
bracis-33630	142	41	size	size	NOUN
bracis-33630	142	42	rate	rate	NOUN
bracis-33630	142	43	(	(	PUNCT
bracis-33630	142	44	ptr	ptr	PROPN
bracis-33630	142	45	)	)	PUNCT
bracis-33630	142	46	assesses	assess	VERB
bracis-33630	142	47	how	how	SCONJ
bracis-33630	142	48	the	the	DET
bracis-33630	142	49	optimization	optimization	NOUN
bracis-33630	142	50	of	of	ADP
bracis-33630	142	51	nmi	nmi	PROPN
bracis-33630	142	52	and	and	CCONJ
bracis-33630	142	53	aic	aic	PROPN
bracis-33630	142	54	affects	affect	VERB
bracis-33630	142	55	the	the	DET
bracis-33630	142	56	ratio	ratio	NOUN
bracis-33630	142	57	between	between	ADP
bracis-33630	142	58	the	the	DET
bracis-33630	142	59	number	number	NOUN
bracis-33630	142	60	of	of	ADP
bracis-33630	142	61	predicted	predict	VERB
bracis-33630	142	62	clusters	cluster	NOUN
bracis-33630	142	63	(	(	PUNCT
bracis-33630	142	64	\(\hat{n}_c\	\(\hat{n}_c\	NUM
bracis-33630	142	65	)	)	PUNCT
bracis-33630	142	66	)	)	PUNCT
bracis-33630	142	67	and	and	CCONJ
bracis-33630	142	68	the	the	DET
bracis-33630	142	69	real	real	ADJ
bracis-33630	142	70	number	number	NOUN
bracis-33630	142	71	of	of	ADP
bracis-33630	142	72	clusters	cluster	NOUN
bracis-33630	142	73	(	(	PUNCT
bracis-33630	142	74	\(n_c\	\(n_c\	NOUN
bracis-33630	142	75	)	)	PUNCT
bracis-33630	142	76	)	)	PUNCT
bracis-33630	142	77	 	 	SPACE
bracis-33630	142	78	and	and	CCONJ
bracis-33630	142	79	is	be	AUX
bracis-33630	142	80	determined	determine	VERB
bracis-33630	142	81	by	by	ADP
bracis-33630	142	82	:	:	PUNCT
bracis-33630	142	83	$	$	SYM
bracis-33630	142	84	$	$	SYM
bracis-33630	142	85	\begin{aligned	\begin{aligne	VERB
bracis-33630	142	86	}	}	PUNCT
bracis-33630	142	87	ptr	ptr	NOUN
bracis-33630	142	88	=	=	SYM
bracis-33630	142	89	\left|	\left|	NOUN
bracis-33630	142	90	\frac{\hat{n}_c}{n_c	\frac{\hat{n}_c}{n_c	NOUN
bracis-33630	142	91	}	}	PUNCT
bracis-33630	142	92	1	1	NUM
bracis-33630	142	93	\right|	\right|	PROPN
bracis-33630	142	94	\end{aligned}$$	\end{aligned}$$	X
bracis-33630	142	95	(	(	PUNCT
bracis-33630	142	96	7	7	NUM
bracis-33630	142	97	)	)	PUNCT
bracis-33630	142	98	these	these	DET
bracis-33630	142	99	metrics	metric	NOUN
bracis-33630	142	100	provide	provide	VERB
bracis-33630	142	101	a	a	DET
bracis-33630	142	102	balanced	balanced	ADJ
bracis-33630	142	103	evaluation	evaluation	NOUN
bracis-33630	142	104	of	of	ADP
bracis-33630	142	105	clustering	clustering	ADJ
bracis-33630	142	106	performance	performance	NOUN
bracis-33630	142	107	and	and	CCONJ
bracis-33630	142	108	model	model	NOUN
bracis-33630	142	109	selection	selection	NOUN
bracis-33630	142	110	,	,	PUNCT
bracis-33630	142	111	ensuring	ensure	VERB
bracis-33630	142	112	robustness	robustness	NOUN
bracis-33630	142	113	in	in	ADP
bracis-33630	142	114	our	our	PRON
bracis-33630	142	115	analysis	analysis	NOUN
bracis-33630	142	116	.	.	PUNCT
bracis-33630	143	1	4	4	NUM
bracis-33630	143	2	experimental	experimental	ADJ
bracis-33630	143	3	study	study	NOUN
bracis-33630	143	4	4.1	4.1	NUM
bracis-33630	143	5	experimental	experimental	ADJ
bracis-33630	143	6	setup	setup	NOUN
bracis-33630	143	7	we	we	PRON
bracis-33630	143	8	evaluated	evaluate	VERB
bracis-33630	143	9	various	various	ADJ
bracis-33630	143	10	techniques	technique	NOUN
bracis-33630	143	11	for	for	ADP
bracis-33630	143	12	clustering	cluster	VERB
bracis-33630	143	13	product	product	NOUN
bracis-33630	143	14	descriptions	description	NOUN
bracis-33630	143	15	,	,	PUNCT
bracis-33630	143	16	focusing	focus	VERB
bracis-33630	143	17	on	on	ADP
bracis-33630	143	18	methods	method	NOUN
bracis-33630	143	19	based	base	VERB
bracis-33630	143	20	on	on	ADP
bracis-33630	143	21	latent	latent	ADJ
bracis-33630	143	22	dirichlet	dirichlet	PROPN
bracis-33630	143	23	allocation	allocation	NOUN
bracis-33630	143	24	(	(	PUNCT
bracis-33630	143	25	lda	lda	PROPN
bracis-33630	143	26	)	)	PUNCT
bracis-33630	143	27	.	.	PUNCT
bracis-33630	144	1	for	for	ADP
bracis-33630	144	2	methods	method	NOUN
bracis-33630	144	3	such	such	ADJ
bracis-33630	144	4	as	as	ADP
bracis-33630	144	5	osdm	osdm	NOUN
bracis-33630	144	6	,	,	PUNCT
bracis-33630	144	7	osgm	osgm	NOUN
bracis-33630	144	8	,	,	PUNCT
bracis-33630	144	9	and	and	CCONJ
bracis-33630	144	10	eindm	eindm	NOUN
bracis-33630	144	11	,	,	PUNCT
bracis-33630	144	12	we	we	PRON
bracis-33630	144	13	carefully	carefully	ADV
bracis-33630	144	14	tuned	tune	VERB
bracis-33630	144	15	the	the	DET
bracis-33630	144	16	lda	lda	PROPN
bracis-33630	144	17	parameters	parameter	NOUN
bracis-33630	144	18	\(\alpha	\(\alpha	X
bracis-33630	144	19	\	\	NOUN
bracis-33630	144	20	)	)	PUNCT
bracis-33630	144	21	(	(	PUNCT
bracis-33630	144	22	document	document	NOUN
bracis-33630	144	23	-	-	PUNCT
bracis-33630	144	24	topic	topic	NOUN
bracis-33630	144	25	distribution	distribution	NOUN
bracis-33630	144	26	)	)	PUNCT
bracis-33630	144	27	and	and	CCONJ
bracis-33630	144	28	\(\beta	\(\beta	PRON
bracis-33630	144	29	\	\	PROPN
bracis-33630	144	30	)	)	PUNCT
bracis-33630	144	31	(	(	PUNCT
bracis-33630	144	32	topic	topic	NOUN
bracis-33630	144	33	-	-	PUNCT
bracis-33630	144	34	word	word	NOUN
bracis-33630	144	35	distribution	distribution	NOUN
bracis-33630	144	36	)	)	PUNCT
bracis-33630	144	37	through	through	ADP
bracis-33630	144	38	a	a	DET
bracis-33630	144	39	grid	grid	NOUN
bracis-33630	144	40	search	search	NOUN
bracis-33630	144	41	to	to	PART
bracis-33630	144	42	optimize	optimize	VERB
bracis-33630	144	43	clustering	clustering	ADJ
bracis-33630	144	44	performance	performance	NOUN
bracis-33630	144	45	.	.	PUNCT
bracis-33630	145	1	additionally	additionally	ADV
bracis-33630	145	2	,	,	PUNCT
bracis-33630	145	3	for	for	ADP
bracis-33630	145	4	these	these	DET
bracis-33630	145	5	methods	method	NOUN
bracis-33630	145	6	,	,	PUNCT
bracis-33630	145	7	the	the	DET
bracis-33630	145	8	decay	decay	NOUN
bracis-33630	145	9	rate	rate	NOUN
bracis-33630	145	10	\(\lambda	\(\lambda	NOUN
bracis-33630	145	11	\	\	NOUN
bracis-33630	145	12	)	)	PUNCT
bracis-33630	145	13	was	be	AUX
bracis-33630	145	14	consistently	consistently	ADV
bracis-33630	145	15	set	set	VERB
bracis-33630	145	16	to	to	ADP
bracis-33630	145	17	\(1\text	\(1\text	NOUN
bracis-33630	145	18	{	{	PUNCT
bracis-33630	145	19	e}{-6}\	e}{-6}\	NOUN
bracis-33630	145	20	)	)	PUNCT
bracis-33630	145	21	,	,	PUNCT
bracis-33630	145	22	with	with	SCONJ
bracis-33630	145	23	all	all	DET
bracis-33630	145	24	other	other	ADJ
bracis-33630	145	25	parameters	parameter	NOUN
bracis-33630	145	26	kept	keep	VERB
bracis-33630	145	27	at	at	ADP
bracis-33630	145	28	their	their	PRON
bracis-33630	145	29	default	default	NOUN
bracis-33630	145	30	values	value	NOUN
bracis-33630	145	31	.	.	PUNCT
bracis-33630	146	1	estream	estream	PROPN
bracis-33630	146	2	requires	require	VERB
bracis-33630	146	3	only	only	ADV
bracis-33630	146	4	one	one	NUM
bracis-33630	146	5	parameter	parameter	NOUN
bracis-33630	146	6	:	:	PUNCT
bracis-33630	146	7	update	update	NOUN
bracis-33630	146	8	-	-	PUNCT
bracis-33630	146	9	interval	interval	NOUN
bracis-33630	146	10	(	(	PUNCT
bracis-33630	146	11	ui	ui	PROPN
bracis-33630	146	12	)	)	PUNCT
bracis-33630	146	13	.	.	PUNCT
bracis-33630	147	1	this	this	DET
bracis-33630	147	2	parameter	parameter	NOUN
bracis-33630	147	3	,	,	PUNCT
bracis-33630	147	4	set	set	VERB
bracis-33630	147	5	to	to	ADP
bracis-33630	147	6	500	500	NUM
bracis-33630	147	7	in	in	ADP
bracis-33630	147	8	our	our	PRON
bracis-33630	147	9	experiments	experiment	NOUN
bracis-33630	147	10	,	,	PUNCT
bracis-33630	147	11	is	be	AUX
bracis-33630	147	12	essential	essential	ADJ
bracis-33630	147	13	for	for	ADP
bracis-33630	147	14	maintaining	maintain	VERB
bracis-33630	147	15	cluster	cluster	NOUN
bracis-33630	147	16	quality	quality	NOUN
bracis-33630	147	17	by	by	ADP
bracis-33630	147	18	updating	update	VERB
bracis-33630	147	19	and	and	CCONJ
bracis-33630	147	20	removing	remove	VERB
bracis-33630	147	21	outdated	outdated	ADJ
bracis-33630	147	22	clusters	cluster	NOUN
bracis-33630	147	23	.	.	PUNCT
bracis-33630	148	1	these	these	DET
bracis-33630	148	2	parameter	parameter	NOUN
bracis-33630	148	3	settings	setting	NOUN
bracis-33630	148	4	were	be	AUX
bracis-33630	148	5	crucial	crucial	ADJ
bracis-33630	148	6	for	for	ADP
bracis-33630	148	7	achieving	achieve	VERB
bracis-33630	148	8	effective	effective	ADJ
bracis-33630	148	9	and	and	CCONJ
bracis-33630	148	10	stable	stable	ADJ
bracis-33630	148	11	clustering	clustering	NOUN
bracis-33630	148	12	results	result	NOUN
bracis-33630	148	13	.	.	PUNCT
bracis-33630	149	1	4.2	4.2	NUM
bracis-33630	149	2	results	result	NOUN
bracis-33630	149	3	in	in	ADP
bracis-33630	149	4	this	this	DET
bracis-33630	149	5	section	section	NOUN
bracis-33630	149	6	,	,	PUNCT
bracis-33630	149	7	we	we	PRON
bracis-33630	149	8	present	present	VERB
bracis-33630	149	9	our	our	PRON
bracis-33630	149	10	findings	finding	NOUN
bracis-33630	149	11	through	through	ADP
bracis-33630	149	12	three	three	NUM
bracis-33630	149	13	primary	primary	ADJ
bracis-33630	149	14	graphs	graph	NOUN
bracis-33630	149	15	.	.	PUNCT
bracis-33630	150	1	the	the	DET
bracis-33630	150	2	initial	initial	ADJ
bracis-33630	150	3	graph	graph	NOUN
bracis-33630	150	4	focuses	focus	VERB
bracis-33630	150	5	on	on	ADP
bracis-33630	150	6	identifying	identify	VERB
bracis-33630	150	7	the	the	DET
bracis-33630	150	8	best	good	ADJ
bracis-33630	150	9	parameter	parameter	NOUN
bracis-33630	150	10	by	by	ADP
bracis-33630	150	11	illustrating	illustrate	VERB
bracis-33630	150	12	the	the	DET
bracis-33630	150	13	performance	performance	NOUN
bracis-33630	150	14	of	of	ADP
bracis-33630	150	15	metrics	metric	NOUN
bracis-33630	150	16	like	like	ADP
bracis-33630	150	17	homogeneity	homogeneity	NOUN
bracis-33630	150	18	,	,	PUNCT
bracis-33630	150	19	completeness	completeness	NOUN
bracis-33630	150	20	,	,	PUNCT
bracis-33630	150	21	nmi	nmi	PROPN
bracis-33630	150	22	,	,	PUNCT
bracis-33630	150	23	purity	purity	NOUN
bracis-33630	150	24	,	,	PUNCT
bracis-33630	150	25	and	and	CCONJ
bracis-33630	150	26	the	the	DET
bracis-33630	150	27	cluster	cluster	NOUN
bracis-33630	150	28	-	-	PUNCT
bracis-33630	150	29	to	to	ADP
bracis-33630	150	30	-	-	PUNCT
bracis-33630	150	31	label	label	NOUN
bracis-33630	150	32	ratio	ratio	NOUN
bracis-33630	150	33	.	.	PUNCT
bracis-33630	151	1	this	this	DET
bracis-33630	151	2	investigation	investigation	NOUN
bracis-33630	151	3	is	be	AUX
bracis-33630	151	4	crucial	crucial	ADJ
bracis-33630	151	5	for	for	ADP
bracis-33630	151	6	determining	determine	VERB
bracis-33630	151	7	the	the	DET
bracis-33630	151	8	conditions	condition	NOUN
bracis-33630	151	9	under	under	ADP
bracis-33630	151	10	which	which	PRON
bracis-33630	151	11	the	the	DET
bracis-33630	151	12	akaike	akaike	ADJ
bracis-33630	151	13	information	information	NOUN
bracis-33630	151	14	criterion	criterion	NOUN
bracis-33630	151	15	(	(	PUNCT
bracis-33630	151	16	aic	aic	PROPN
bracis-33630	151	17	)	)	PUNCT
bracis-33630	151	18	is	be	AUX
bracis-33630	151	19	minimized	minimize	VERB
bracis-33630	151	20	,	,	PUNCT
bracis-33630	151	21	as	as	SCONJ
bracis-33630	151	22	shown	show	VERB
bracis-33630	151	23	in	in	ADP
bracis-33630	151	24	fig	fig	NOUN
bracis-33630	151	25	.	.	PUNCT
bracis-33630	151	26	 	 	SPACE
bracis-33630	152	1	2	2	X
bracis-33630	152	2	.	.	PUNCT
bracis-33630	153	1	our	our	PRON
bracis-33630	153	2	objective	objective	NOUN
bracis-33630	153	3	is	be	AUX
bracis-33630	153	4	to	to	PART
bracis-33630	153	5	reveal	reveal	VERB
bracis-33630	153	6	how	how	SCONJ
bracis-33630	153	7	different	different	ADJ
bracis-33630	153	8	settings	setting	NOUN
bracis-33630	153	9	affect	affect	VERB
bracis-33630	153	10	clustering	cluster	VERB
bracis-33630	153	11	quality	quality	NOUN
bracis-33630	153	12	,	,	PUNCT
bracis-33630	153	13	mainly	mainly	ADV
bracis-33630	153	14	through	through	ADP
bracis-33630	153	15	the	the	DET
bracis-33630	153	16	lens	lens	NOUN
bracis-33630	153	17	of	of	ADP
bracis-33630	153	18	aic	aic	PROPN
bracis-33630	153	19	optimization	optimization	PROPN
bracis-33630	153	20	.	.	PUNCT
bracis-33630	154	1	fig	fig	NOUN
bracis-33630	154	2	.	.	PUNCT
bracis-33630	155	1	2	2	X
bracis-33630	155	2	.	.	X
bracis-33630	155	3	result	result	NOUN
bracis-33630	155	4	of	of	ADP
bracis-33630	155	5	homogeneity	homogeneity	NOUN
bracis-33630	155	6	,	,	PUNCT
bracis-33630	155	7	completeness	completeness	NOUN
bracis-33630	155	8	,	,	PUNCT
bracis-33630	155	9	purity	purity	NOUN
bracis-33630	155	10	,	,	PUNCT
bracis-33630	155	11	nmi	nmi	PROPN
bracis-33630	155	12	,	,	PUNCT
bracis-33630	155	13	predict	predict	VERB
bracis-33630	155	14	/	/	SYM
bracis-33630	155	15	true	true	ADJ
bracis-33630	155	16	size	size	NOUN
bracis-33630	155	17	rate	rate	NOUN
bracis-33630	155	18	,	,	PUNCT
bracis-33630	155	19	and	and	CCONJ
bracis-33630	155	20	aic	aic	PROPN
bracis-33630	155	21	.	.	PROPN
bracis-33630	156	1	for	for	ADP
bracis-33630	156	2	gtin	gtin	NOUN
bracis-33630	156	3	,	,	PUNCT
bracis-33630	156	4	ncm4	ncm4	PROPN
bracis-33630	156	5	,	,	PUNCT
bracis-33630	156	6	and	and	CCONJ
bracis-33630	156	7	ncm8	ncm8	PROPN
bracis-33630	156	8	datasets	dataset	NOUN
bracis-33630	156	9	in	in	ADP
bracis-33630	156	10	best	good	ADJ
bracis-33630	156	11	nmi	nmi	PROPN
bracis-33630	156	12	analyses	analysis	NOUN
bracis-33630	156	13	context	context	NOUN
bracis-33630	156	14	.	.	PUNCT
bracis-33630	157	1	full	full	ADJ
bracis-33630	157	2	size	size	NOUN
bracis-33630	157	3	image	image	NOUN
bracis-33630	157	4	the	the	DET
bracis-33630	157	5	second	second	ADJ
bracis-33630	157	6	graph	graph	NOUN
bracis-33630	157	7	illustrates	illustrate	VERB
bracis-33630	157	8	the	the	DET
bracis-33630	157	9	time	time	NOUN
bracis-33630	157	10	evolution	evolution	NOUN
bracis-33630	157	11	for	for	ADP
bracis-33630	157	12	four	four	NUM
bracis-33630	157	13	methods	method	NOUN
bracis-33630	157	14	across	across	ADP
bracis-33630	157	15	varying	vary	VERB
bracis-33630	157	16	dataset	dataset	ADJ
bracis-33630	157	17	sizes	size	NOUN
bracis-33630	157	18	,	,	PUNCT
bracis-33630	157	19	as	as	SCONJ
bracis-33630	157	20	depicted	depict	VERB
bracis-33630	157	21	in	in	ADP
bracis-33630	157	22	fig	fig	NOUN
bracis-33630	157	23	.	.	PUNCT
bracis-33630	157	24	 	 	SPACE
bracis-33630	158	1	3	3	X
bracis-33630	158	2	.	.	PUNCT
bracis-33630	159	1	each	each	DET
bracis-33630	159	2	method	method	NOUN
bracis-33630	159	3	’s	’s	PART
bracis-33630	159	4	processing	processing	NOUN
bracis-33630	159	5	time	time	NOUN
bracis-33630	159	6	,	,	PUNCT
bracis-33630	159	7	represented	represent	VERB
bracis-33630	159	8	in	in	ADP
bracis-33630	159	9	seconds	second	NOUN
bracis-33630	159	10	,	,	PUNCT
bracis-33630	159	11	is	be	AUX
bracis-33630	159	12	plotted	plot	VERB
bracis-33630	159	13	against	against	ADP
bracis-33630	159	14	the	the	DET
bracis-33630	159	15	dataset	dataset	NOUN
bracis-33630	159	16	size	size	NOUN
bracis-33630	159	17	.	.	PUNCT
bracis-33630	160	1	the	the	DET
bracis-33630	160	2	logarithmic	logarithmic	ADJ
bracis-33630	160	3	scale	scale	NOUN
bracis-33630	160	4	on	on	ADP
bracis-33630	160	5	the	the	DET
bracis-33630	160	6	y	y	PROPN
bracis-33630	160	7	-	-	PUNCT
bracis-33630	160	8	axis	axis	NOUN
bracis-33630	160	9	facilitates	facilitate	VERB
bracis-33630	160	10	comparing	compare	VERB
bracis-33630	160	11	relative	relative	ADJ
bracis-33630	160	12	changes	change	NOUN
bracis-33630	160	13	in	in	ADP
bracis-33630	160	14	processing	processing	NOUN
bracis-33630	160	15	time	time	NOUN
bracis-33630	160	16	.	.	PUNCT
bracis-33630	161	1	this	this	DET
bracis-33630	161	2	visualization	visualization	NOUN
bracis-33630	161	3	offers	offer	VERB
bracis-33630	161	4	insights	insight	NOUN
bracis-33630	161	5	into	into	ADP
bracis-33630	161	6	how	how	SCONJ
bracis-33630	161	7	each	each	DET
bracis-33630	161	8	method	method	NOUN
bracis-33630	161	9	’s	’s	PART
bracis-33630	161	10	processing	processing	NOUN
bracis-33630	161	11	time	time	NOUN
bracis-33630	161	12	scales	scale	VERB
bracis-33630	161	13	with	with	ADP
bracis-33630	161	14	dataset	dataset	ADJ
bracis-33630	161	15	size	size	NOUN
bracis-33630	161	16	,	,	PUNCT
bracis-33630	161	17	providing	provide	VERB
bracis-33630	161	18	valuable	valuable	ADJ
bracis-33630	161	19	information	information	NOUN
bracis-33630	161	20	on	on	ADP
bracis-33630	161	21	computational	computational	ADJ
bracis-33630	161	22	efficiency	efficiency	NOUN
bracis-33630	161	23	and	and	CCONJ
bracis-33630	161	24	scalability	scalability	NOUN
bracis-33630	161	25	without	without	ADP
bracis-33630	161	26	delving	delve	VERB
bracis-33630	161	27	into	into	ADP
bracis-33630	161	28	specific	specific	ADJ
bracis-33630	161	29	numeral	numeral	ADJ
bracis-33630	161	30	evaluations	evaluation	NOUN
bracis-33630	161	31	.	.	PUNCT
bracis-33630	162	1	fig	fig	NOUN
bracis-33630	162	2	.	.	PUNCT
bracis-33630	163	1	3	3	X
bracis-33630	163	2	.	.	NOUN
bracis-33630	163	3	time	time	NOUN
bracis-33630	163	4	evolution	evolution	NOUN
bracis-33630	163	5	(	(	PUNCT
bracis-33630	163	6	processing	processing	NOUN
bracis-33630	163	7	time	time	NOUN
bracis-33630	163	8	in	in	ADP
bracis-33630	163	9	seconds	second	NOUN
bracis-33630	163	10	)	)	PUNCT
bracis-33630	163	11	versus	versus	ADP
bracis-33630	163	12	the	the	DET
bracis-33630	163	13	size	size	NOUN
bracis-33630	163	14	of	of	ADP
bracis-33630	163	15	the	the	DET
bracis-33630	163	16	gtin	gtin	NOUN
bracis-33630	163	17	dataset	dataset	VERB
bracis-33630	163	18	for	for	ADP
bracis-33630	163	19	four	four	NUM
bracis-33630	163	20	different	different	ADJ
bracis-33630	163	21	methods	method	NOUN
bracis-33630	163	22	:	:	PUNCT
bracis-33630	163	23	eindm	eindm	NOUN
bracis-33630	163	24	,	,	PUNCT
bracis-33630	163	25	osgm	osgm	NOUN
bracis-33630	163	26	,	,	PUNCT
bracis-33630	163	27	osdm	osdm	NOUN
bracis-33630	163	28	,	,	PUNCT
bracis-33630	163	29	and	and	CCONJ
bracis-33630	163	30	estream	estream	NOUN
bracis-33630	163	31	.	.	PUNCT
bracis-33630	164	1	full	full	ADJ
bracis-33630	164	2	size	size	NOUN
bracis-33630	164	3	image	image	NOUN
bracis-33630	164	4	the	the	DET
bracis-33630	164	5	third	third	ADJ
bracis-33630	164	6	graph	graph	NOUN
bracis-33630	164	7	demonstrates	demonstrate	VERB
bracis-33630	164	8	the	the	DET
bracis-33630	164	9	evolution	evolution	NOUN
bracis-33630	164	10	of	of	ADP
bracis-33630	164	11	the	the	DET
bracis-33630	164	12	\(\alpha	\(\alpha	ADJ
bracis-33630	164	13	\	\	NOUN
bracis-33630	164	14	)	)	PUNCT
bracis-33630	164	15	and	and	CCONJ
bracis-33630	164	16	\(\beta	\(\beta	PRON
bracis-33630	164	17	\	\	ADJ
bracis-33630	164	18	)	)	PUNCT
bracis-33630	164	19	parameters	parameter	NOUN
bracis-33630	164	20	,	,	PUNCT
bracis-33630	164	21	as	as	SCONJ
bracis-33630	164	22	depicted	depict	VERB
bracis-33630	164	23	in	in	ADP
bracis-33630	164	24	fig	fig	NOUN
bracis-33630	164	25	.	.	PUNCT
bracis-33630	164	26	 	 	SPACE
bracis-33630	165	1	4	4	X
bracis-33630	165	2	.	.	PUNCT
bracis-33630	166	1	this	this	DET
bracis-33630	166	2	analysis	analysis	NOUN
bracis-33630	166	3	segment	segment	NOUN
bracis-33630	166	4	is	be	AUX
bracis-33630	166	5	essential	essential	ADJ
bracis-33630	166	6	for	for	ADP
bracis-33630	166	7	comprehending	comprehend	VERB
bracis-33630	166	8	how	how	SCONJ
bracis-33630	166	9	nmi	nmi	PROPN
bracis-33630	166	10	and	and	CCONJ
bracis-33630	166	11	aic	aic	PROPN
bracis-33630	166	12	adjust	adjust	VERB
bracis-33630	166	13	to	to	ADP
bracis-33630	166	14	parameter	parameter	NOUN
bracis-33630	166	15	changes	change	NOUN
bracis-33630	166	16	and	and	CCONJ
bracis-33630	166	17	for	for	ADP
bracis-33630	166	18	identifying	identify	VERB
bracis-33630	166	19	the	the	DET
bracis-33630	166	20	configurations	configuration	NOUN
bracis-33630	166	21	that	that	PRON
bracis-33630	166	22	lead	lead	VERB
bracis-33630	166	23	to	to	ADP
bracis-33630	166	24	the	the	DET
bracis-33630	166	25	best	good	ADJ
bracis-33630	166	26	results	result	NOUN
bracis-33630	166	27	.	.	PUNCT
bracis-33630	167	1	the	the	DET
bracis-33630	167	2	study	study	NOUN
bracis-33630	167	3	focuses	focus	VERB
bracis-33630	167	4	on	on	ADP
bracis-33630	167	5	the	the	DET
bracis-33630	167	6	methods	method	NOUN
bracis-33630	167	7	eindm	eindm	NOUN
bracis-33630	167	8	,	,	PUNCT
bracis-33630	167	9	osdm	osdm	NOUN
bracis-33630	167	10	,	,	PUNCT
bracis-33630	167	11	and	and	CCONJ
bracis-33630	167	12	osgm	osgm	NOUN
bracis-33630	167	13	while	while	SCONJ
bracis-33630	167	14	omitting	omit	VERB
bracis-33630	167	15	estream	estream	NOUN
bracis-33630	167	16	since	since	SCONJ
bracis-33630	167	17	it	it	PRON
bracis-33630	167	18	does	do	AUX
bracis-33630	167	19	not	not	PART
bracis-33630	167	20	involve	involve	VERB
bracis-33630	167	21	parametric	parametric	ADJ
bracis-33630	167	22	techniques	technique	NOUN
bracis-33630	167	23	.	.	PUNCT
bracis-33630	168	1	this	this	DET
bracis-33630	168	2	focused	focus	VERB
bracis-33630	168	3	examination	examination	NOUN
bracis-33630	168	4	aids	aid	NOUN
bracis-33630	168	5	in	in	ADP
bracis-33630	168	6	identifying	identify	VERB
bracis-33630	168	7	the	the	DET
bracis-33630	168	8	optimal	optimal	ADJ
bracis-33630	168	9	parameters	parameter	NOUN
bracis-33630	168	10	for	for	ADP
bracis-33630	168	11	enhanced	enhanced	ADJ
bracis-33630	168	12	clustering	clustering	NOUN
bracis-33630	168	13	efficiency	efficiency	NOUN
bracis-33630	168	14	.	.	PUNCT
bracis-33630	169	1	fig	fig	NOUN
bracis-33630	169	2	.	.	PUNCT
bracis-33630	170	1	4	4	X
bracis-33630	170	2	.	.	PUNCT
bracis-33630	170	3	\(\alpha	\(\alpha	PROPN
bracis-33630	170	4	\	\	NOUN
bracis-33630	170	5	)	)	PUNCT
bracis-33630	170	6	and	and	CCONJ
bracis-33630	170	7	\(\beta	\(\beta	PRON
bracis-33630	170	8	\	\	ADJ
bracis-33630	170	9	)	)	PUNCT
bracis-33630	170	10	evolution	evolution	NOUN
bracis-33630	170	11	for	for	ADP
bracis-33630	170	12	the	the	DET
bracis-33630	170	13	methods	method	NOUN
bracis-33630	170	14	eindm	eindm	NOUN
bracis-33630	170	15	,	,	PUNCT
bracis-33630	170	16	osdm	osdm	NOUN
bracis-33630	170	17	and	and	CCONJ
bracis-33630	170	18	osgm	osgm	NOUN
bracis-33630	170	19	.	.	PUNCT
bracis-33630	171	1	full	full	ADJ
bracis-33630	171	2	size	size	NOUN
bracis-33630	171	3	image	image	NOUN
bracis-33630	171	4	5	5	NUM
bracis-33630	171	5	discussion	discussion	NOUN
bracis-33630	171	6	aic	aic	PROPN
bracis-33630	171	7	analysis	analysis	NOUN
bracis-33630	171	8	highlights	highlight	NOUN
bracis-33630	171	9	cb	cb	PROPN
bracis-33630	171	10	-	-	PUNCT
bracis-33630	171	11	tmcoh	tmcoh	NOUN
bracis-33630	171	12	and	and	CCONJ
bracis-33630	171	13	tmcoh	tmcoh	NOUN
bracis-33630	171	14	’s	’s	PART
bracis-33630	171	15	superior	superior	ADJ
bracis-33630	171	16	efficiency	efficiency	NOUN
bracis-33630	171	17	compared	compare	VERB
bracis-33630	171	18	to	to	ADP
bracis-33630	171	19	mstream	mstream	VERB
bracis-33630	171	20	,	,	PUNCT
bracis-33630	171	21	indicating	indicate	VERB
bracis-33630	171	22	better	well	ADJ
bracis-33630	171	23	optimization	optimization	NOUN
bracis-33630	171	24	of	of	ADP
bracis-33630	171	25	cluster	cluster	NOUN
bracis-33630	171	26	numbers	number	NOUN
bracis-33630	171	27	without	without	ADP
bracis-33630	171	28	sacrificing	sacrifice	VERB
bracis-33630	171	29	model	model	NOUN
bracis-33630	171	30	complexity	complexity	NOUN
bracis-33630	171	31	.	.	PUNCT
bracis-33630	172	1	this	this	PRON
bracis-33630	172	2	underscores	underscore	VERB
bracis-33630	172	3	their	their	PRON
bracis-33630	172	4	ability	ability	NOUN
bracis-33630	172	5	to	to	PART
bracis-33630	172	6	capture	capture	VERB
bracis-33630	172	7	data	data	NOUN
bracis-33630	172	8	structure	structure	NOUN
bracis-33630	172	9	effectively	effectively	ADV
bracis-33630	172	10	while	while	SCONJ
bracis-33630	172	11	avoiding	avoid	VERB
bracis-33630	172	12	overfitting	overfitte	VERB
bracis-33630	172	13	.	.	PUNCT
bracis-33630	173	1	additionally	additionally	ADV
bracis-33630	173	2	,	,	PUNCT
bracis-33630	173	3	cb	cb	PROPN
bracis-33630	173	4	-	-	PUNCT
bracis-33630	173	5	tmcoh	tmcoh	NOUN
bracis-33630	173	6	outperforms	outperform	NOUN
bracis-33630	173	7	existing	exist	VERB
bracis-33630	173	8	lda	lda	PROPN
bracis-33630	173	9	methods	method	NOUN
bracis-33630	173	10	like	like	ADP
bracis-33630	173	11	tmcoh	tmcoh	NOUN
bracis-33630	173	12	and	and	CCONJ
bracis-33630	173	13	faststream	faststream	NOUN
bracis-33630	173	14	,	,	PUNCT
bracis-33630	173	15	particularly	particularly	ADV
bracis-33630	173	16	regarding	regard	VERB
bracis-33630	173	17	completeness	completeness	NOUN
bracis-33630	173	18	and	and	CCONJ
bracis-33630	173	19	purity	purity	NOUN
bracis-33630	173	20	.	.	PUNCT
bracis-33630	174	1	this	this	PRON
bracis-33630	174	2	demonstrates	demonstrate	VERB
bracis-33630	174	3	its	its	PRON
bracis-33630	174	4	effectiveness	effectiveness	NOUN
bracis-33630	174	5	in	in	ADP
bracis-33630	174	6	grouping	group	VERB
bracis-33630	174	7	similar	similar	ADJ
bracis-33630	174	8	elements	element	NOUN
bracis-33630	174	9	while	while	SCONJ
bracis-33630	174	10	ensuring	ensure	VERB
bracis-33630	174	11	class	class	NOUN
bracis-33630	174	12	exclusivity	exclusivity	NOUN
bracis-33630	174	13	within	within	ADP
bracis-33630	174	14	clusters	cluster	NOUN
bracis-33630	174	15	.	.	PUNCT
bracis-33630	175	1	5.1	5.1	NUM
bracis-33630	175	2	metrics	metric	NOUN
bracis-33630	175	3	evaluation	evaluation	NOUN
bracis-33630	175	4	for	for	ADP
bracis-33630	175	5	datasets	dataset	NOUN
bracis-33630	175	6	ncm4	ncm4	PROPN
bracis-33630	175	7	,	,	PUNCT
bracis-33630	175	8	ncm8	ncm8	PROPN
bracis-33630	175	9	,	,	PUNCT
bracis-33630	175	10	and	and	CCONJ
bracis-33630	175	11	gtin	gtin	NOUN
bracis-33630	175	12	,	,	PUNCT
bracis-33630	175	13	homogeneity	homogeneity	NOUN
bracis-33630	175	14	varies	vary	VERB
bracis-33630	175	15	significantly	significantly	ADV
bracis-33630	175	16	.	.	PUNCT
bracis-33630	176	1	in	in	ADP
bracis-33630	176	2	ncm4	ncm4	PROPN
bracis-33630	176	3	,	,	PUNCT
bracis-33630	176	4	with	with	ADP
bracis-33630	176	5	fewer	few	ADJ
bracis-33630	176	6	clusters	cluster	NOUN
bracis-33630	176	7	,	,	PUNCT
bracis-33630	176	8	homogeneity	homogeneity	NOUN
bracis-33630	176	9	is	be	AUX
bracis-33630	176	10	generally	generally	ADV
bracis-33630	176	11	higher	high	ADJ
bracis-33630	176	12	as	as	SCONJ
bracis-33630	176	13	clusters	cluster	NOUN
bracis-33630	176	14	are	be	AUX
bracis-33630	176	15	likely	likely	ADV
bracis-33630	176	16	more	more	ADV
bracis-33630	176	17	uniform	uniform	ADJ
bracis-33630	176	18	but	but	CCONJ
bracis-33630	176	19	could	could	AUX
bracis-33630	176	20	be	be	AUX
bracis-33630	176	21	less	less	ADV
bracis-33630	176	22	comprehensive	comprehensive	ADJ
bracis-33630	176	23	.	.	PUNCT
bracis-33630	177	1	in	in	ADP
bracis-33630	177	2	contrast	contrast	NOUN
bracis-33630	177	3	,	,	PUNCT
bracis-33630	177	4	ncm8	ncm8	PROPN
bracis-33630	177	5	and	and	CCONJ
bracis-33630	177	6	gtin	gtin	PROPN
bracis-33630	177	7	,	,	PUNCT
bracis-33630	177	8	with	with	ADP
bracis-33630	177	9	more	more	ADV
bracis-33630	177	10	complex	complex	ADJ
bracis-33630	177	11	and	and	CCONJ
bracis-33630	177	12	numerous	numerous	ADJ
bracis-33630	177	13	clusters	cluster	NOUN
bracis-33630	177	14	,	,	PUNCT
bracis-33630	177	15	show	show	VERB
bracis-33630	177	16	a	a	DET
bracis-33630	177	17	potential	potential	ADJ
bracis-33630	177	18	decrease	decrease	NOUN
bracis-33630	177	19	in	in	ADP
bracis-33630	177	20	homogeneity	homogeneity	NOUN
bracis-33630	177	21	due	due	ADP
bracis-33630	177	22	to	to	ADP
bracis-33630	177	23	the	the	DET
bracis-33630	177	24	wider	wide	ADJ
bracis-33630	177	25	variety	variety	NOUN
bracis-33630	177	26	of	of	ADP
bracis-33630	177	27	data	datum	NOUN
bracis-33630	177	28	points	point	NOUN
bracis-33630	177	29	within	within	ADP
bracis-33630	177	30	each	each	DET
bracis-33630	177	31	cluster	cluster	NOUN
bracis-33630	177	32	.	.	PUNCT
bracis-33630	178	1	comparatively	comparatively	ADV
bracis-33630	178	2	,	,	PUNCT
bracis-33630	178	3	between	between	ADP
bracis-33630	178	4	the	the	DET
bracis-33630	178	5	two	two	NUM
bracis-33630	178	6	cases	case	NOUN
bracis-33630	178	7	(	(	PUNCT
bracis-33630	178	8	e.g.	e.g.	ADV
bracis-33630	178	9	,	,	PUNCT
bracis-33630	178	10	best	good	ADJ
bracis-33630	178	11	aic	aic	PROPN
bracis-33630	178	12	vs	vs	ADP
bracis-33630	178	13	best	good	ADJ
bracis-33630	178	14	nmi	nmi	PROPN
bracis-33630	178	15	)	)	PUNCT
bracis-33630	178	16	,	,	PUNCT
bracis-33630	178	17	best	good	ADJ
bracis-33630	178	18	nmi	nmi	NOUN
bracis-33630	178	19	prioritizes	prioritize	VERB
bracis-33630	178	20	achieving	achieve	VERB
bracis-33630	178	21	higher	high	ADJ
bracis-33630	178	22	homogeneity	homogeneity	NOUN
bracis-33630	178	23	as	as	SCONJ
bracis-33630	178	24	it	it	PRON
bracis-33630	178	25	directly	directly	ADV
bracis-33630	178	26	influences	influence	VERB
bracis-33630	178	27	the	the	DET
bracis-33630	178	28	mutual	mutual	ADJ
bracis-33630	178	29	information	information	NOUN
bracis-33630	178	30	score	score	NOUN
bracis-33630	178	31	,	,	PUNCT
bracis-33630	178	32	leading	lead	VERB
bracis-33630	178	33	to	to	ADP
bracis-33630	178	34	better	well	ADV
bracis-33630	178	35	-	-	PUNCT
bracis-33630	178	36	aligning	align	VERB
bracis-33630	178	37	clusters	cluster	NOUN
bracis-33630	178	38	with	with	ADP
bracis-33630	178	39	actual	actual	ADJ
bracis-33630	178	40	data	datum	NOUN
bracis-33630	178	41	classes	class	NOUN
bracis-33630	178	42	.	.	PUNCT
bracis-33630	179	1	ncm4	ncm4	PROPN
bracis-33630	179	2	’s	’s	PART
bracis-33630	179	3	completeness	completeness	NOUN
bracis-33630	179	4	tends	tend	VERB
bracis-33630	179	5	to	to	PART
bracis-33630	179	6	be	be	AUX
bracis-33630	179	7	high	high	ADJ
bracis-33630	179	8	due	due	ADJ
bracis-33630	179	9	to	to	ADP
bracis-33630	179	10	simpler	simple	ADJ
bracis-33630	179	11	data	datum	NOUN
bracis-33630	179	12	structures	structure	NOUN
bracis-33630	179	13	,	,	PUNCT
bracis-33630	179	14	where	where	SCONJ
bracis-33630	179	15	capturing	capture	VERB
bracis-33630	179	16	all	all	DET
bracis-33630	179	17	class	class	NOUN
bracis-33630	179	18	members	member	NOUN
bracis-33630	179	19	in	in	ADP
bracis-33630	179	20	a	a	DET
bracis-33630	179	21	single	single	ADJ
bracis-33630	179	22	cluster	cluster	NOUN
bracis-33630	179	23	is	be	AUX
bracis-33630	179	24	easier	easy	ADJ
bracis-33630	179	25	.	.	PUNCT
bracis-33630	180	1	however	however	ADV
bracis-33630	180	2	,	,	PUNCT
bracis-33630	180	3	in	in	ADP
bracis-33630	180	4	ncm8	ncm8	PROPN
bracis-33630	180	5	and	and	CCONJ
bracis-33630	180	6	gtin	gtin	NOUN
bracis-33630	180	7	,	,	PUNCT
bracis-33630	180	8	increasing	increase	VERB
bracis-33630	180	9	data	datum	NOUN
bracis-33630	180	10	complexity	complexity	NOUN
bracis-33630	180	11	and	and	CCONJ
bracis-33630	180	12	cluster	cluster	NOUN
bracis-33630	180	13	numbers	number	NOUN
bracis-33630	180	14	can	can	AUX
bracis-33630	180	15	reduce	reduce	VERB
bracis-33630	180	16	completeness	completeness	NOUN
bracis-33630	180	17	as	as	SCONJ
bracis-33630	180	18	classes	class	NOUN
bracis-33630	180	19	may	may	AUX
bracis-33630	180	20	spread	spread	VERB
bracis-33630	180	21	over	over	ADP
bracis-33630	180	22	multiple	multiple	ADJ
bracis-33630	180	23	clusters	cluster	NOUN
bracis-33630	180	24	.	.	PUNCT
bracis-33630	181	1	between	between	ADP
bracis-33630	181	2	cases	case	NOUN
bracis-33630	181	3	,	,	PUNCT
bracis-33630	181	4	best	good	ADJ
bracis-33630	181	5	aic	aic	PROPN
bracis-33630	181	6	might	might	AUX
bracis-33630	181	7	show	show	VERB
bracis-33630	181	8	reduced	reduced	ADJ
bracis-33630	181	9	completeness	completeness	NOUN
bracis-33630	181	10	compared	compare	VERB
bracis-33630	181	11	to	to	ADP
bracis-33630	181	12	best	good	ADJ
bracis-33630	181	13	nmi	nmi	PROPN
bracis-33630	181	14	because	because	SCONJ
bracis-33630	181	15	aic	aic	PROPN
bracis-33630	181	16	focuses	focus	VERB
bracis-33630	181	17	more	more	ADJ
bracis-33630	181	18	on	on	ADP
bracis-33630	181	19	model	model	NOUN
bracis-33630	181	20	simplicity	simplicity	NOUN
bracis-33630	181	21	and	and	CCONJ
bracis-33630	181	22	goodness	goodness	NOUN
bracis-33630	181	23	of	of	ADP
bracis-33630	181	24	fit	fit	NOUN
bracis-33630	181	25	rather	rather	ADV
bracis-33630	181	26	than	than	ADP
bracis-33630	181	27	ensuring	ensure	VERB
bracis-33630	181	28	all	all	DET
bracis-33630	181	29	class	class	NOUN
bracis-33630	181	30	members	member	NOUN
bracis-33630	181	31	are	be	AUX
bracis-33630	181	32	together	together	ADV
bracis-33630	181	33	,	,	PUNCT
bracis-33630	181	34	directly	directly	ADV
bracis-33630	181	35	influencing	influence	VERB
bracis-33630	181	36	nmi	nmi	PROPN
bracis-33630	181	37	.	.	PUNCT
bracis-33630	182	1	purity	purity	NOUN
bracis-33630	182	2	is	be	AUX
bracis-33630	182	3	generally	generally	ADV
bracis-33630	182	4	higher	high	ADJ
bracis-33630	182	5	in	in	ADP
bracis-33630	182	6	datasets	dataset	NOUN
bracis-33630	182	7	with	with	ADP
bracis-33630	182	8	fewer	few	ADJ
bracis-33630	182	9	clusters	cluster	NOUN
bracis-33630	182	10	like	like	ADP
bracis-33630	182	11	ncm4	ncm4	PROPN
bracis-33630	182	12	,	,	PUNCT
bracis-33630	182	13	where	where	SCONJ
bracis-33630	182	14	the	the	DET
bracis-33630	182	15	likelihood	likelihood	NOUN
bracis-33630	182	16	of	of	ADP
bracis-33630	182	17	mixing	mix	VERB
bracis-33630	182	18	different	different	ADJ
bracis-33630	182	19	classes	class	NOUN
bracis-33630	182	20	in	in	ADP
bracis-33630	182	21	one	one	NUM
bracis-33630	182	22	cluster	cluster	NOUN
bracis-33630	182	23	is	be	AUX
bracis-33630	182	24	lower	low	ADJ
bracis-33630	182	25	.	.	PUNCT
bracis-33630	183	1	in	in	ADP
bracis-33630	183	2	datasets	dataset	NOUN
bracis-33630	183	3	like	like	ADP
bracis-33630	183	4	ncm8	ncm8	NOUN
bracis-33630	183	5	and	and	CCONJ
bracis-33630	183	6	gtin	gtin	NOUN
bracis-33630	183	7	,	,	PUNCT
bracis-33630	183	8	despite	despite	SCONJ
bracis-33630	183	9	a	a	DET
bracis-33630	183	10	higher	high	ADJ
bracis-33630	183	11	number	number	NOUN
bracis-33630	183	12	of	of	ADP
bracis-33630	183	13	clusters	cluster	NOUN
bracis-33630	183	14	,	,	PUNCT
bracis-33630	183	15	the	the	DET
bracis-33630	183	16	complexity	complexity	NOUN
bracis-33630	183	17	of	of	ADP
bracis-33630	183	18	data	datum	NOUN
bracis-33630	183	19	might	might	AUX
bracis-33630	183	20	lead	lead	VERB
bracis-33630	183	21	to	to	ADP
bracis-33630	183	22	lower	low	ADJ
bracis-33630	183	23	purity	purity	NOUN
bracis-33630	183	24	as	as	SCONJ
bracis-33630	183	25	clusters	cluster	NOUN
bracis-33630	183	26	can	can	AUX
bracis-33630	183	27	contain	contain	VERB
bracis-33630	183	28	a	a	DET
bracis-33630	183	29	more	more	ADV
bracis-33630	183	30	diverse	diverse	ADJ
bracis-33630	183	31	set	set	NOUN
bracis-33630	183	32	of	of	ADP
bracis-33630	183	33	class	class	NOUN
bracis-33630	183	34	labels	label	NOUN
bracis-33630	183	35	.	.	PUNCT
bracis-33630	184	1	in	in	ADP
bracis-33630	184	2	terms	term	NOUN
bracis-33630	184	3	of	of	ADP
bracis-33630	184	4	cases	case	NOUN
bracis-33630	184	5	,	,	PUNCT
bracis-33630	184	6	best	good	ADJ
bracis-33630	184	7	nmi	nmi	PROPN
bracis-33630	184	8	likely	likely	ADV
bracis-33630	184	9	shows	show	VERB
bracis-33630	184	10	better	well	ADJ
bracis-33630	184	11	purity	purity	NOUN
bracis-33630	184	12	than	than	ADP
bracis-33630	184	13	best	good	ADJ
bracis-33630	184	14	aic	aic	PROPN
bracis-33630	184	15	,	,	PUNCT
bracis-33630	184	16	as	as	SCONJ
bracis-33630	184	17	maximizing	maximize	VERB
bracis-33630	184	18	mutual	mutual	ADJ
bracis-33630	184	19	information	information	NOUN
bracis-33630	184	20	would	would	AUX
bracis-33630	184	21	inadvertently	inadvertently	ADV
bracis-33630	184	22	align	align	VERB
bracis-33630	184	23	clusters	cluster	NOUN
bracis-33630	184	24	more	more	ADV
bracis-33630	184	25	cleanly	cleanly	ADV
bracis-33630	184	26	with	with	ADP
bracis-33630	184	27	true	true	ADJ
bracis-33630	184	28	class	class	NOUN
bracis-33630	184	29	labels	label	NOUN
bracis-33630	184	30	.	.	PUNCT
bracis-33630	185	1	nmi	nmi	PROPN
bracis-33630	185	2	is	be	AUX
bracis-33630	185	3	particularly	particularly	ADV
bracis-33630	185	4	insightful	insightful	ADJ
bracis-33630	185	5	for	for	ADP
bracis-33630	185	6	comparing	compare	VERB
bracis-33630	185	7	clustering	clustering	ADJ
bracis-33630	185	8	performance	performance	NOUN
bracis-33630	185	9	across	across	ADP
bracis-33630	185	10	datasets	dataset	NOUN
bracis-33630	185	11	.	.	PUNCT
bracis-33630	186	1	ncm4	ncm4	PROPN
bracis-33630	186	2	might	might	AUX
bracis-33630	186	3	show	show	VERB
bracis-33630	186	4	stable	stable	ADJ
bracis-33630	186	5	nmi	nmi	PROPN
bracis-33630	186	6	values	value	NOUN
bracis-33630	186	7	due	due	ADJ
bracis-33630	186	8	to	to	ADP
bracis-33630	186	9	less	less	ADJ
bracis-33630	186	10	complexity	complexity	NOUN
bracis-33630	186	11	,	,	PUNCT
bracis-33630	186	12	while	while	SCONJ
bracis-33630	186	13	ncm8	ncm8	PROPN
bracis-33630	186	14	and	and	CCONJ
bracis-33630	186	15	gtin	gtin	NOUN
bracis-33630	186	16	could	could	AUX
bracis-33630	186	17	exhibit	exhibit	VERB
bracis-33630	186	18	more	more	ADJ
bracis-33630	186	19	variability	variability	NOUN
bracis-33630	186	20	in	in	ADP
bracis-33630	186	21	nmi	nmi	PROPN
bracis-33630	186	22	due	due	ADP
bracis-33630	186	23	to	to	ADP
bracis-33630	186	24	their	their	PRON
bracis-33630	186	25	increased	increase	VERB
bracis-33630	186	26	complexity	complexity	NOUN
bracis-33630	186	27	and	and	CCONJ
bracis-33630	186	28	cluster	cluster	NOUN
bracis-33630	186	29	counts	count	NOUN
bracis-33630	186	30	.	.	PUNCT
bracis-33630	187	1	comparing	compare	VERB
bracis-33630	187	2	cases	case	NOUN
bracis-33630	187	3	,	,	PUNCT
bracis-33630	187	4	best	good	ADJ
bracis-33630	187	5	nmi	nmi	PROPN
bracis-33630	187	6	is	be	AUX
bracis-33630	187	7	explicitly	explicitly	ADV
bracis-33630	187	8	optimized	optimize	VERB
bracis-33630	187	9	to	to	PART
bracis-33630	187	10	maximize	maximize	VERB
bracis-33630	187	11	this	this	DET
bracis-33630	187	12	metric	metric	ADJ
bracis-33630	187	13	,	,	PUNCT
bracis-33630	187	14	likely	likely	ADV
bracis-33630	187	15	resulting	result	VERB
bracis-33630	187	16	in	in	ADP
bracis-33630	187	17	higher	high	ADJ
bracis-33630	187	18	nmi	nmi	PROPN
bracis-33630	187	19	scores	score	NOUN
bracis-33630	187	20	across	across	ADP
bracis-33630	187	21	all	all	DET
bracis-33630	187	22	datasets	dataset	NOUN
bracis-33630	187	23	than	than	ADP
bracis-33630	187	24	best	good	ADJ
bracis-33630	187	25	aic	aic	PROPN
bracis-33630	187	26	,	,	PUNCT
bracis-33630	187	27	which	which	PRON
bracis-33630	187	28	might	might	AUX
bracis-33630	187	29	sacrifice	sacrifice	VERB
bracis-33630	187	30	some	some	DET
bracis-33630	187	31	mutual	mutual	ADJ
bracis-33630	187	32	information	information	NOUN
bracis-33630	187	33	for	for	ADP
bracis-33630	187	34	simpler	simple	ADJ
bracis-33630	187	35	model	model	NOUN
bracis-33630	187	36	structures	structure	NOUN
bracis-33630	187	37	.	.	PUNCT
bracis-33630	188	1	the	the	DET
bracis-33630	188	2	ptr	ptr	PROPN
bracis-33630	188	3	measures	measure	VERB
bracis-33630	188	4	the	the	DET
bracis-33630	188	5	clustering	clustering	ADJ
bracis-33630	188	6	algorithm	algorithm	NOUN
bracis-33630	188	7	’s	’s	PART
bracis-33630	188	8	efficiency	efficiency	NOUN
bracis-33630	188	9	in	in	ADP
bracis-33630	188	10	controlling	control	VERB
bracis-33630	188	11	cluster	cluster	NOUN
bracis-33630	188	12	sizes	size	NOUN
bracis-33630	188	13	,	,	PUNCT
bracis-33630	188	14	ideally	ideally	ADV
bracis-33630	188	15	approaching	approach	VERB
bracis-33630	188	16	1	1	NUM
bracis-33630	188	17	.	.	PUNCT
bracis-33630	189	1	ncm4	ncm4	PROPN
bracis-33630	189	2	usually	usually	ADV
bracis-33630	189	3	shows	show	VERB
bracis-33630	189	4	higher	high	ADJ
bracis-33630	189	5	ptr	ptr	NOUN
bracis-33630	189	6	,	,	PUNCT
bracis-33630	189	7	indicating	indicate	VERB
bracis-33630	189	8	over	over	ADP
bracis-33630	189	9	-	-	PUNCT
bracis-33630	189	10	clustering	clustering	NOUN
bracis-33630	189	11	.	.	PUNCT
bracis-33630	190	1	ncm8	ncm8	PROPN
bracis-33630	190	2	and	and	CCONJ
bracis-33630	190	3	gtin	gtin	PROPN
bracis-33630	190	4	show	show	VERB
bracis-33630	190	5	lower	low	ADJ
bracis-33630	190	6	ptr	ptr	NOUN
bracis-33630	190	7	due	due	ADP
bracis-33630	190	8	to	to	ADP
bracis-33630	190	9	their	their	PRON
bracis-33630	190	10	complex	complex	ADJ
bracis-33630	190	11	structures	structure	NOUN
bracis-33630	190	12	,	,	PUNCT
bracis-33630	190	13	leading	lead	VERB
bracis-33630	190	14	to	to	ADP
bracis-33630	190	15	many	many	ADJ
bracis-33630	190	16	small	small	ADJ
bracis-33630	190	17	clusters	cluster	NOUN
bracis-33630	190	18	.	.	PUNCT
bracis-33630	191	1	best	good	ADJ
bracis-33630	191	2	aic	aic	PROPN
bracis-33630	191	3	,	,	PUNCT
bracis-33630	191	4	favoring	favor	VERB
bracis-33630	191	5	simpler	simple	ADJ
bracis-33630	191	6	models	model	NOUN
bracis-33630	191	7	,	,	PUNCT
bracis-33630	191	8	typically	typically	ADV
bracis-33630	191	9	has	have	VERB
bracis-33630	191	10	ptr	ptr	PROPN
bracis-33630	191	11	closer	close	ADV
bracis-33630	191	12	to	to	ADP
bracis-33630	191	13	1	1	NUM
bracis-33630	191	14	than	than	ADP
bracis-33630	191	15	best	good	ADJ
bracis-33630	191	16	nmi	nmi	PROPN
bracis-33630	191	17	.	.	PUNCT
bracis-33630	192	1	best	good	ADJ
bracis-33630	192	2	nmi	nmi	PROPN
bracis-33630	192	3	might	might	AUX
bracis-33630	192	4	increase	increase	VERB
bracis-33630	192	5	the	the	DET
bracis-33630	192	6	number	number	NOUN
bracis-33630	192	7	of	of	ADP
bracis-33630	192	8	clusters	cluster	NOUN
bracis-33630	192	9	to	to	PART
bracis-33630	192	10	boost	boost	VERB
bracis-33630	192	11	mutual	mutual	ADJ
bracis-33630	192	12	information	information	NOUN
bracis-33630	192	13	,	,	PUNCT
bracis-33630	192	14	resulting	result	VERB
bracis-33630	192	15	in	in	ADP
bracis-33630	192	16	lower	low	ADJ
bracis-33630	192	17	ptr	ptr	NOUN
bracis-33630	192	18	.	.	PUNCT
bracis-33630	193	1	aic	aic	PROPN
bracis-33630	193	2	values	value	NOUN
bracis-33630	193	3	can	can	AUX
bracis-33630	193	4	help	help	VERB
bracis-33630	193	5	gauge	gauge	VERB
bracis-33630	193	6	the	the	DET
bracis-33630	193	7	efficacy	efficacy	NOUN
bracis-33630	193	8	of	of	ADP
bracis-33630	193	9	clustering	cluster	VERB
bracis-33630	193	10	concerning	concern	VERB
bracis-33630	193	11	model	model	NOUN
bracis-33630	193	12	simplicity	simplicity	NOUN
bracis-33630	193	13	.	.	PUNCT
bracis-33630	194	1	lower	low	ADJ
bracis-33630	194	2	aic	aic	PROPN
bracis-33630	194	3	values	value	NOUN
bracis-33630	194	4	in	in	ADP
bracis-33630	194	5	ncm4	ncm4	PROPN
bracis-33630	194	6	suggest	suggest	VERB
bracis-33630	194	7	a	a	DET
bracis-33630	194	8	better	well	ADJ
bracis-33630	194	9	fit	fit	NOUN
bracis-33630	194	10	with	with	ADP
bracis-33630	194	11	simpler	simple	ADJ
bracis-33630	194	12	models	model	NOUN
bracis-33630	194	13	.	.	PUNCT
bracis-33630	195	1	in	in	ADP
bracis-33630	195	2	contrast	contrast	NOUN
bracis-33630	195	3	,	,	PUNCT
bracis-33630	195	4	higher	high	ADJ
bracis-33630	195	5	aic	aic	PROPN
bracis-33630	195	6	in	in	ADP
bracis-33630	195	7	ncm8	ncm8	PROPN
bracis-33630	195	8	and	and	CCONJ
bracis-33630	195	9	gtin	gtin	NOUN
bracis-33630	195	10	might	might	AUX
bracis-33630	195	11	indicate	indicate	VERB
bracis-33630	195	12	that	that	SCONJ
bracis-33630	195	13	the	the	DET
bracis-33630	195	14	models	model	NOUN
bracis-33630	195	15	,	,	PUNCT
bracis-33630	195	16	despite	despite	SCONJ
bracis-33630	195	17	being	be	AUX
bracis-33630	195	18	more	more	ADV
bracis-33630	195	19	complex	complex	ADJ
bracis-33630	195	20	,	,	PUNCT
bracis-33630	195	21	are	be	AUX
bracis-33630	195	22	necessary	necessary	ADJ
bracis-33630	195	23	to	to	PART
bracis-33630	195	24	fit	fit	VERB
bracis-33630	195	25	the	the	DET
bracis-33630	195	26	data	datum	NOUN
bracis-33630	195	27	adequately	adequately	ADV
bracis-33630	195	28	.	.	PUNCT
bracis-33630	196	1	comparatively	comparatively	ADV
bracis-33630	196	2	,	,	PUNCT
bracis-33630	196	3	best	good	ADJ
bracis-33630	196	4	aic	aic	PROPN
bracis-33630	196	5	consistently	consistently	ADV
bracis-33630	196	6	focuses	focus	VERB
bracis-33630	196	7	on	on	ADP
bracis-33630	196	8	minimizing	minimize	VERB
bracis-33630	196	9	aic	aic	PROPN
bracis-33630	196	10	,	,	PUNCT
bracis-33630	196	11	potentially	potentially	ADV
bracis-33630	196	12	leading	lead	VERB
bracis-33630	196	13	to	to	ADP
bracis-33630	196	14	simpler	simple	ADJ
bracis-33630	196	15	,	,	PUNCT
bracis-33630	196	16	more	more	ADV
bracis-33630	196	17	adaptable	adaptable	ADJ
bracis-33630	196	18	models	model	NOUN
bracis-33630	196	19	across	across	ADP
bracis-33630	196	20	datasets	dataset	NOUN
bracis-33630	196	21	than	than	ADP
bracis-33630	196	22	best	good	ADJ
bracis-33630	196	23	nmi	nmi	PROPN
bracis-33630	196	24	,	,	PUNCT
bracis-33630	196	25	which	which	PRON
bracis-33630	196	26	might	might	AUX
bracis-33630	196	27	accept	accept	VERB
bracis-33630	196	28	more	more	ADJ
bracis-33630	196	29	complex	complex	ADJ
bracis-33630	196	30	models	model	NOUN
bracis-33630	196	31	if	if	SCONJ
bracis-33630	196	32	they	they	PRON
bracis-33630	196	33	yield	yield	VERB
bracis-33630	196	34	better	well	ADJ
bracis-33630	196	35	information	information	NOUN
bracis-33630	196	36	sharing	sharing	NOUN
bracis-33630	196	37	.	.	PUNCT
bracis-33630	197	1	5.2	5.2	NUM
bracis-33630	197	2	assessment	assessment	NOUN
bracis-33630	197	3	of	of	ADP
bracis-33630	197	4	 	 	SPACE
bracis-33630	197	5	methods	method	NOUN
bracis-33630	197	6	eindm	eindm	NOUN
bracis-33630	197	7	is	be	AUX
bracis-33630	197	8	likely	likely	ADV
bracis-33630	197	9	effective	effective	ADJ
bracis-33630	197	10	in	in	ADP
bracis-33630	197	11	datasets	dataset	NOUN
bracis-33630	197	12	like	like	ADP
bracis-33630	197	13	gtin	gtin	NOUN
bracis-33630	197	14	,	,	PUNCT
bracis-33630	197	15	where	where	SCONJ
bracis-33630	197	16	data	datum	NOUN
bracis-33630	197	17	may	may	AUX
bracis-33630	197	18	be	be	AUX
bracis-33630	197	19	diverse	diverse	ADJ
bracis-33630	197	20	and	and	CCONJ
bracis-33630	197	21	voluminous	voluminous	ADJ
bracis-33630	197	22	.	.	PUNCT
bracis-33630	198	1	eindm	eindm	PROPN
bracis-33630	198	2	can	can	AUX
bracis-33630	198	3	adapt	adapt	VERB
bracis-33630	198	4	to	to	ADP
bracis-33630	198	5	complex	complex	ADJ
bracis-33630	198	6	structures	structure	NOUN
bracis-33630	198	7	by	by	ADP
bracis-33630	198	8	combining	combine	VERB
bracis-33630	198	9	multiple	multiple	ADJ
bracis-33630	198	10	models	model	NOUN
bracis-33630	198	11	to	to	PART
bracis-33630	198	12	improve	improve	VERB
bracis-33630	198	13	clustering	clustering	ADJ
bracis-33630	198	14	accuracy	accuracy	NOUN
bracis-33630	198	15	.	.	PUNCT
bracis-33630	199	1	in	in	ADP
bracis-33630	199	2	ncm4	ncm4	PROPN
bracis-33630	199	3	,	,	PUNCT
bracis-33630	199	4	which	which	PRON
bracis-33630	199	5	may	may	AUX
bracis-33630	199	6	be	be	AUX
bracis-33630	199	7	simpler	simple	ADJ
bracis-33630	199	8	,	,	PUNCT
bracis-33630	199	9	the	the	DET
bracis-33630	199	10	ensemble	ensemble	ADJ
bracis-33630	199	11	approach	approach	NOUN
bracis-33630	199	12	might	might	AUX
bracis-33630	199	13	not	not	PART
bracis-33630	199	14	significantly	significantly	ADV
bracis-33630	199	15	outperform	outperform	VERB
bracis-33630	199	16	simpler	simple	ADJ
bracis-33630	199	17	methods	method	NOUN
bracis-33630	199	18	but	but	CCONJ
bracis-33630	199	19	could	could	AUX
bracis-33630	199	20	still	still	ADV
bracis-33630	199	21	provide	provide	VERB
bracis-33630	199	22	robustness	robustness	NOUN
bracis-33630	199	23	against	against	ADP
bracis-33630	199	24	noise	noise	NOUN
bracis-33630	199	25	.	.	PUNCT
bracis-33630	200	1	in	in	ADP
bracis-33630	200	2	best	good	ADJ
bracis-33630	200	3	aic	aic	PROPN
bracis-33630	200	4	,	,	PUNCT
bracis-33630	200	5	eindm	eindm	PROPN
bracis-33630	200	6	might	might	AUX
bracis-33630	200	7	be	be	AUX
bracis-33630	200	8	configured	configure	VERB
bracis-33630	200	9	between	between	ADP
bracis-33630	200	10	the	the	DET
bracis-33630	200	11	two	two	NUM
bracis-33630	200	12	cases	case	NOUN
bracis-33630	200	13	to	to	PART
bracis-33630	200	14	limit	limit	VERB
bracis-33630	200	15	the	the	DET
bracis-33630	200	16	model	model	NOUN
bracis-33630	200	17	complexity	complexity	NOUN
bracis-33630	200	18	and	and	CCONJ
bracis-33630	200	19	avoid	avoid	VERB
bracis-33630	200	20	unnecessary	unnecessary	ADJ
bracis-33630	200	21	computation	computation	NOUN
bracis-33630	200	22	.	.	PUNCT
bracis-33630	201	1	in	in	ADP
bracis-33630	201	2	contrast	contrast	NOUN
bracis-33630	201	3	,	,	PUNCT
bracis-33630	201	4	in	in	ADP
bracis-33630	201	5	best	good	ADJ
bracis-33630	201	6	nmi	nmi	NOUN
bracis-33630	201	7	,	,	PUNCT
bracis-33630	201	8	the	the	DET
bracis-33630	201	9	ensemble	ensemble	ADJ
bracis-33630	201	10	could	could	AUX
bracis-33630	201	11	be	be	AUX
bracis-33630	201	12	tuned	tune	VERB
bracis-33630	201	13	to	to	PART
bracis-33630	201	14	maximize	maximize	VERB
bracis-33630	201	15	information	information	NOUN
bracis-33630	201	16	captured	capture	VERB
bracis-33630	201	17	from	from	ADP
bracis-33630	201	18	the	the	DET
bracis-33630	201	19	data	datum	NOUN
bracis-33630	201	20	,	,	PUNCT
bracis-33630	201	21	potentially	potentially	ADV
bracis-33630	201	22	at	at	ADP
bracis-33630	201	23	the	the	DET
bracis-33630	201	24	expense	expense	NOUN
bracis-33630	201	25	of	of	ADP
bracis-33630	201	26	increased	increase	VERB
bracis-33630	201	27	computational	computational	ADJ
bracis-33630	201	28	load	load	NOUN
bracis-33630	201	29	.	.	PUNCT
bracis-33630	202	1	estream	estream	PROPN
bracis-33630	202	2	tends	tend	VERB
bracis-33630	202	3	to	to	PART
bracis-33630	202	4	form	form	VERB
bracis-33630	202	5	more	more	ADJ
bracis-33630	202	6	clusters	cluster	NOUN
bracis-33630	202	7	than	than	SCONJ
bracis-33630	202	8	exist	exist	VERB
bracis-33630	202	9	,	,	PUNCT
bracis-33630	202	10	performing	perform	VERB
bracis-33630	202	11	better	well	ADV
bracis-33630	202	12	only	only	ADV
bracis-33630	202	13	than	than	ADP
bracis-33630	202	14	osgm	osgm	NOUN
bracis-33630	202	15	.	.	PUNCT
bracis-33630	203	1	it	it	PRON
bracis-33630	203	2	achieves	achieve	VERB
bracis-33630	203	3	worse	bad	ADJ
bracis-33630	203	4	results	result	NOUN
bracis-33630	203	5	than	than	ADP
bracis-33630	203	6	the	the	DET
bracis-33630	203	7	best	good	ADJ
bracis-33630	203	8	nmi	nmi	NOUN
bracis-33630	203	9	cases	case	NOUN
bracis-33630	203	10	but	but	CCONJ
bracis-33630	203	11	better	well	ADJ
bracis-33630	203	12	than	than	ADP
bracis-33630	203	13	the	the	DET
bracis-33630	203	14	best	good	ADJ
bracis-33630	203	15	aic	aic	PROPN
bracis-33630	203	16	cases	case	NOUN
bracis-33630	203	17	.	.	PUNCT
bracis-33630	204	1	however	however	ADV
bracis-33630	204	2	,	,	PUNCT
bracis-33630	204	3	it	it	PRON
bracis-33630	204	4	offers	offer	VERB
bracis-33630	204	5	an	an	DET
bracis-33630	204	6	excellent	excellent	ADJ
bracis-33630	204	7	cost	cost	NOUN
bracis-33630	204	8	-	-	PUNCT
bracis-33630	204	9	benefit	benefit	NOUN
bracis-33630	204	10	ratio	ratio	NOUN
bracis-33630	204	11	as	as	SCONJ
bracis-33630	204	12	it	it	PRON
bracis-33630	204	13	is	be	AUX
bracis-33630	204	14	the	the	DET
bracis-33630	204	15	most	most	ADV
bracis-33630	204	16	computationally	computationally	ADV
bracis-33630	204	17	efficient	efficient	ADJ
bracis-33630	204	18	method	method	NOUN
bracis-33630	204	19	,	,	PUNCT
bracis-33630	204	20	especially	especially	ADV
bracis-33630	204	21	for	for	SCONJ
bracis-33630	204	22	the	the	DET
bracis-33630	204	23	gtin	gtin	NOUN
bracis-33630	204	24	dataset	dataset	VERB
bracis-33630	204	25	.	.	PUNCT
bracis-33630	205	1	odsm	odsm	PROPN
bracis-33630	205	2	achieves	achieve	VERB
bracis-33630	205	3	its	its	PRON
bracis-33630	205	4	best	good	ADJ
bracis-33630	205	5	performance	performance	NOUN
bracis-33630	205	6	with	with	ADP
bracis-33630	205	7	smaller	small	ADJ
bracis-33630	205	8	and	and	CCONJ
bracis-33630	205	9	simpler	simple	ADJ
bracis-33630	205	10	datasets	dataset	NOUN
bracis-33630	205	11	like	like	ADP
bracis-33630	205	12	ncm4	ncm4	PROPN
bracis-33630	205	13	.	.	PUNCT
bracis-33630	206	1	as	as	ADP
bracis-33630	206	2	the	the	DET
bracis-33630	206	3	complexity	complexity	NOUN
bracis-33630	206	4	of	of	ADP
bracis-33630	206	5	the	the	DET
bracis-33630	206	6	dataset	dataset	NOUN
bracis-33630	206	7	increases	increase	NOUN
bracis-33630	206	8	,	,	PUNCT
bracis-33630	206	9	its	its	PRON
bracis-33630	206	10	performance	performance	NOUN
bracis-33630	206	11	degrades	degrade	VERB
bracis-33630	206	12	.	.	PUNCT
bracis-33630	207	1	odsm	odsm	NOUN
bracis-33630	207	2	does	do	AUX
bracis-33630	207	3	not	not	PART
bracis-33630	207	4	tend	tend	VERB
bracis-33630	207	5	to	to	PART
bracis-33630	207	6	form	form	VERB
bracis-33630	207	7	many	many	ADJ
bracis-33630	207	8	clusters	cluster	NOUN
bracis-33630	207	9	for	for	ADP
bracis-33630	207	10	datasets	dataset	NOUN
bracis-33630	207	11	with	with	ADP
bracis-33630	207	12	many	many	ADJ
bracis-33630	207	13	classes	class	NOUN
bracis-33630	207	14	,	,	PUNCT
bracis-33630	207	15	maintaining	maintain	VERB
bracis-33630	207	16	a	a	DET
bracis-33630	207	17	predict	predict	ADJ
bracis-33630	207	18	/	/	SYM
bracis-33630	207	19	true	true	ADJ
bracis-33630	207	20	size	size	NOUN
bracis-33630	207	21	rate	rate	NOUN
bracis-33630	207	22	close	close	ADJ
bracis-33630	207	23	to	to	PART
bracis-33630	207	24	ideal	ideal	ADJ
bracis-33630	207	25	.	.	PUNCT
bracis-33630	208	1	osgm	osgm	PROPN
bracis-33630	208	2	has	have	VERB
bracis-33630	208	3	the	the	DET
bracis-33630	208	4	worst	bad	ADJ
bracis-33630	208	5	case	case	NOUN
bracis-33630	208	6	for	for	ADP
bracis-33630	208	7	aic	aic	PROPN
bracis-33630	208	8	values	value	NOUN
bracis-33630	208	9	due	due	ADP
bracis-33630	208	10	to	to	ADP
bracis-33630	208	11	its	its	PRON
bracis-33630	208	12	tendency	tendency	NOUN
bracis-33630	208	13	to	to	PART
bracis-33630	208	14	form	form	VERB
bracis-33630	208	15	many	many	ADJ
bracis-33630	208	16	clusters	cluster	NOUN
bracis-33630	208	17	.	.	PUNCT
bracis-33630	209	1	even	even	ADV
bracis-33630	209	2	in	in	ADP
bracis-33630	209	3	the	the	DET
bracis-33630	209	4	best	good	ADJ
bracis-33630	209	5	aic	aic	PROPN
bracis-33630	209	6	case	case	NOUN
bracis-33630	209	7	,	,	PUNCT
bracis-33630	209	8	this	this	DET
bracis-33630	209	9	method	method	NOUN
bracis-33630	209	10	generates	generate	VERB
bracis-33630	209	11	more	more	ADJ
bracis-33630	209	12	clusters	cluster	NOUN
bracis-33630	209	13	than	than	ADP
bracis-33630	209	14	others	other	NOUN
bracis-33630	209	15	.	.	PUNCT
bracis-33630	210	1	the	the	DET
bracis-33630	210	2	best	good	ADJ
bracis-33630	210	3	scenario	scenario	NOUN
bracis-33630	210	4	to	to	PART
bracis-33630	210	5	use	use	VERB
bracis-33630	210	6	osgm	osgm	NOUN
bracis-33630	210	7	is	be	AUX
bracis-33630	210	8	with	with	SCONJ
bracis-33630	210	9	the	the	DET
bracis-33630	210	10	gtin	gtin	NOUN
bracis-33630	210	11	dataset	dataset	VERB
bracis-33630	210	12	and	and	CCONJ
bracis-33630	210	13	with	with	ADP
bracis-33630	210	14	large	large	ADJ
bracis-33630	210	15	datasets	dataset	NOUN
bracis-33630	210	16	.	.	PUNCT
bracis-33630	211	1	between	between	ADP
bracis-33630	211	2	cases	case	NOUN
bracis-33630	211	3	,	,	PUNCT
bracis-33630	211	4	best	good	ADJ
bracis-33630	211	5	aic	aic	PROPN
bracis-33630	211	6	would	would	AUX
bracis-33630	211	7	prefer	prefer	VERB
bracis-33630	211	8	setting	set	VERB
bracis-33630	211	9	stringent	stringent	ADJ
bracis-33630	211	10	criteria	criterion	NOUN
bracis-33630	211	11	for	for	ADP
bracis-33630	211	12	model	model	NOUN
bracis-33630	211	13	updates	update	NOUN
bracis-33630	211	14	to	to	PART
bracis-33630	211	15	avoid	avoid	VERB
bracis-33630	211	16	complexity	complexity	NOUN
bracis-33630	211	17	.	.	PUNCT
bracis-33630	212	1	at	at	ADP
bracis-33630	212	2	the	the	DET
bracis-33630	212	3	same	same	ADJ
bracis-33630	212	4	time	time	NOUN
bracis-33630	212	5	,	,	PUNCT
bracis-33630	212	6	best	good	ADJ
bracis-33630	212	7	nmi	nmi	PROPN
bracis-33630	212	8	might	might	AUX
bracis-33630	212	9	leverage	leverage	VERB
bracis-33630	212	10	the	the	DET
bracis-33630	212	11	flexibility	flexibility	NOUN
bracis-33630	212	12	of	of	ADP
bracis-33630	212	13	gaussian	gaussian	ADJ
bracis-33630	212	14	mixtures	mixture	NOUN
bracis-33630	212	15	to	to	PART
bracis-33630	212	16	refine	refine	VERB
bracis-33630	212	17	cluster	cluster	NOUN
bracis-33630	212	18	definitions	definition	NOUN
bracis-33630	212	19	continuously	continuously	ADV
bracis-33630	212	20	,	,	PUNCT
bracis-33630	212	21	aiming	aim	VERB
bracis-33630	212	22	for	for	ADP
bracis-33630	212	23	higher	high	ADJ
bracis-33630	212	24	nmi	nmi	PROPN
bracis-33630	212	25	scores	score	NOUN
bracis-33630	212	26	.	.	PUNCT
bracis-33630	213	1	each	each	DET
bracis-33630	213	2	method	method	NOUN
bracis-33630	213	3	exhibits	exhibit	VERB
bracis-33630	213	4	a	a	DET
bracis-33630	213	5	distinct	distinct	ADJ
bracis-33630	213	6	pattern	pattern	NOUN
bracis-33630	213	7	of	of	ADP
bracis-33630	213	8	processing	processing	NOUN
bracis-33630	213	9	time	time	NOUN
bracis-33630	213	10	as	as	ADP
bracis-33630	213	11	dataset	dataset	ADJ
bracis-33630	213	12	size	size	NOUN
bracis-33630	213	13	increases	increase	NOUN
bracis-33630	213	14	.	.	PUNCT
bracis-33630	214	1	eindm	eindm	PROPN
bracis-33630	214	2	demonstrates	demonstrate	VERB
bracis-33630	214	3	a	a	DET
bracis-33630	214	4	notable	notable	ADJ
bracis-33630	214	5	increase	increase	NOUN
bracis-33630	214	6	in	in	ADP
bracis-33630	214	7	processing	processing	NOUN
bracis-33630	214	8	time	time	NOUN
bracis-33630	214	9	with	with	ADP
bracis-33630	214	10	larger	large	ADJ
bracis-33630	214	11	datasets	dataset	NOUN
bracis-33630	214	12	,	,	PUNCT
bracis-33630	214	13	while	while	SCONJ
bracis-33630	214	14	osgm	osgm	NOUN
bracis-33630	214	15	and	and	CCONJ
bracis-33630	214	16	osdm	osdm	NOUN
bracis-33630	214	17	also	also	ADV
bracis-33630	214	18	show	show	VERB
bracis-33630	214	19	an	an	DET
bracis-33630	214	20	upward	upward	ADJ
bracis-33630	214	21	trend	trend	NOUN
bracis-33630	214	22	,	,	PUNCT
bracis-33630	214	23	albeit	albeit	SCONJ
bracis-33630	214	24	with	with	ADP
bracis-33630	214	25	different	different	ADJ
bracis-33630	214	26	slopes	slope	NOUN
bracis-33630	214	27	.	.	PUNCT
bracis-33630	215	1	in	in	ADP
bracis-33630	215	2	contrast	contrast	NOUN
bracis-33630	215	3	,	,	PUNCT
bracis-33630	215	4	estream	estream	NOUN
bracis-33630	215	5	exhibits	exhibit	VERB
bracis-33630	215	6	significantly	significantly	ADV
bracis-33630	215	7	lower	low	ADJ
bracis-33630	215	8	processing	processing	NOUN
bracis-33630	215	9	times	time	NOUN
bracis-33630	215	10	across	across	ADP
bracis-33630	215	11	all	all	DET
bracis-33630	215	12	dataset	dataset	ADJ
bracis-33630	215	13	sizes	size	NOUN
bracis-33630	215	14	,	,	PUNCT
bracis-33630	215	15	indicating	indicate	VERB
bracis-33630	215	16	its	its	PRON
bracis-33630	215	17	efficiency	efficiency	NOUN
bracis-33630	215	18	in	in	ADP
bracis-33630	215	19	handling	handle	VERB
bracis-33630	215	20	larger	large	ADJ
bracis-33630	215	21	datasets	dataset	NOUN
bracis-33630	215	22	.	.	PUNCT
bracis-33630	216	1	5.3	5.3	NUM
bracis-33630	216	2	hyperparameters	hyperparameter	VERB
bracis-33630	216	3	the	the	DET
bracis-33630	216	4	parameter	parameter	NOUN
bracis-33630	216	5	\(\alpha	\(\alpha	NOUN
bracis-33630	216	6	\	\	NOUN
bracis-33630	216	7	)	)	PUNCT
bracis-33630	216	8	shows	show	VERB
bracis-33630	216	9	a	a	DET
bracis-33630	216	10	varying	vary	VERB
bracis-33630	216	11	pattern	pattern	NOUN
bracis-33630	216	12	across	across	ADP
bracis-33630	216	13	the	the	DET
bracis-33630	216	14	dataset	dataset	NOUN
bracis-33630	216	15	sizes	size	NOUN
bracis-33630	216	16	for	for	ADP
bracis-33630	216	17	the	the	DET
bracis-33630	216	18	eindm	eindm	NOUN
bracis-33630	216	19	method	method	NOUN
bracis-33630	216	20	.	.	PUNCT
bracis-33630	217	1	at	at	ADP
bracis-33630	217	2	a	a	DET
bracis-33630	217	3	size	size	NOUN
bracis-33630	217	4	of	of	ADP
bracis-33630	217	5	10,000	10,000	NUM
bracis-33630	217	6	,	,	PUNCT
bracis-33630	217	7	the	the	DET
bracis-33630	217	8	\(\alpha	\(\alpha	NOUN
bracis-33630	217	9	\	\	NOUN
bracis-33630	217	10	)	)	PUNCT
bracis-33630	217	11	values	value	NOUN
bracis-33630	217	12	are	be	AUX
bracis-33630	217	13	0.4	0.4	NUM
bracis-33630	217	14	and	and	CCONJ
bracis-33630	217	15	1.0	1.0	NUM
bracis-33630	217	16	for	for	ADP
bracis-33630	217	17	cases	case	NOUN
bracis-33630	217	18	best	well	ADV
bracis-33630	217	19	aic	aic	PROPN
bracis-33630	217	20	and	and	CCONJ
bracis-33630	217	21	best	good	ADJ
bracis-33630	217	22	nmi	nmi	PROPN
bracis-33630	217	23	,	,	PUNCT
bracis-33630	217	24	respectively	respectively	ADV
bracis-33630	217	25	.	.	PUNCT
bracis-33630	218	1	as	as	SCONJ
bracis-33630	218	2	the	the	DET
bracis-33630	218	3	dataset	dataset	NOUN
bracis-33630	218	4	size	size	NOUN
bracis-33630	218	5	increases	increase	VERB
bracis-33630	218	6	to	to	ADP
bracis-33630	218	7	20,000	20,000	NUM
bracis-33630	218	8	and	and	CCONJ
bracis-33630	218	9	30,000	30,000	NUM
bracis-33630	218	10	,	,	PUNCT
bracis-33630	218	11	the	the	DET
bracis-33630	218	12	\(\alpha	\(\alpha	NOUN
bracis-33630	218	13	\	\	NOUN
bracis-33630	218	14	)	)	PUNCT
bracis-33630	218	15	values	value	NOUN
bracis-33630	218	16	remain	remain	VERB
bracis-33630	218	17	consistent	consistent	ADJ
bracis-33630	218	18	at	at	ADP
bracis-33630	218	19	0.4	0.4	NUM
bracis-33630	218	20	for	for	ADP
bracis-33630	218	21	the	the	DET
bracis-33630	218	22	best	good	ADJ
bracis-33630	218	23	aic	aic	PROPN
bracis-33630	218	24	case	case	NOUN
bracis-33630	218	25	.	.	PUNCT
bracis-33630	219	1	still	still	ADV
bracis-33630	219	2	,	,	PUNCT
bracis-33630	219	3	the	the	DET
bracis-33630	219	4	values	value	NOUN
bracis-33630	219	5	adjust	adjust	VERB
bracis-33630	219	6	slightly	slightly	ADV
bracis-33630	219	7	to	to	ADP
bracis-33630	219	8	1.0	1.0	NUM
bracis-33630	219	9	for	for	ADP
bracis-33630	219	10	the	the	DET
bracis-33630	219	11	best	good	ADJ
bracis-33630	219	12	nmi	nmi	PROPN
bracis-33630	219	13	case	case	NOUN
bracis-33630	219	14	,	,	PUNCT
bracis-33630	219	15	indicating	indicate	VERB
bracis-33630	219	16	a	a	DET
bracis-33630	219	17	stable	stable	ADJ
bracis-33630	219	18	but	but	CCONJ
bracis-33630	219	19	distinct	distinct	ADJ
bracis-33630	219	20	configuration	configuration	NOUN
bracis-33630	219	21	for	for	ADP
bracis-33630	219	22	optimal	optimal	ADJ
bracis-33630	219	23	clustering	clustering	NOUN
bracis-33630	219	24	performance	performance	NOUN
bracis-33630	219	25	.	.	PUNCT
bracis-33630	220	1	this	this	DET
bracis-33630	220	2	evolution	evolution	NOUN
bracis-33630	220	3	suggests	suggest	VERB
bracis-33630	220	4	that	that	SCONJ
bracis-33630	220	5	the	the	DET
bracis-33630	220	6	\(\alpha	\(\alpha	NOUN
bracis-33630	220	7	\	\	NOUN
bracis-33630	220	8	)	)	PUNCT
bracis-33630	220	9	parameter	parameter	NOUN
bracis-33630	220	10	is	be	AUX
bracis-33630	220	11	finely	finely	ADV
bracis-33630	220	12	tuned	tune	VERB
bracis-33630	220	13	to	to	PART
bracis-33630	220	14	balance	balance	VERB
bracis-33630	220	15	,	,	PUNCT
bracis-33630	220	16	preserving	preserve	VERB
bracis-33630	220	17	cluster	cluster	NOUN
bracis-33630	220	18	structure	structure	NOUN
bracis-33630	220	19	and	and	CCONJ
bracis-33630	220	20	enhancing	enhance	VERB
bracis-33630	220	21	cluster	cluster	NOUN
bracis-33630	220	22	homogeneity	homogeneity	NOUN
bracis-33630	220	23	.	.	PUNCT
bracis-33630	221	1	the	the	DET
bracis-33630	221	2	parameter	parameter	NOUN
bracis-33630	221	3	\(\beta	\(\beta	X
bracis-33630	221	4	\	\	PROPN
bracis-33630	221	5	)	)	PUNCT
bracis-33630	221	6	exhibits	exhibit	VERB
bracis-33630	221	7	a	a	DET
bracis-33630	221	8	steady	steady	ADJ
bracis-33630	221	9	behaviour	behaviour	NOUN
bracis-33630	221	10	,	,	PUNCT
bracis-33630	221	11	particularly	particularly	ADV
bracis-33630	221	12	for	for	ADP
bracis-33630	221	13	the	the	DET
bracis-33630	221	14	best	good	ADJ
bracis-33630	221	15	aic	aic	PROPN
bracis-33630	221	16	cases	case	NOUN
bracis-33630	221	17	,	,	PUNCT
bracis-33630	221	18	where	where	SCONJ
bracis-33630	221	19	it	it	PRON
bracis-33630	221	20	remains	remain	VERB
bracis-33630	221	21	constant	constant	ADJ
bracis-33630	221	22	at	at	ADP
bracis-33630	221	23	0.200	0.200	NUM
bracis-33630	221	24	for	for	ADP
bracis-33630	221	25	the	the	DET
bracis-33630	221	26	initial	initial	ADJ
bracis-33630	221	27	dataset	dataset	ADJ
bracis-33630	221	28	sizes	size	NOUN
bracis-33630	221	29	of	of	ADP
bracis-33630	221	30	10,000	10,000	NUM
bracis-33630	221	31	and	and	CCONJ
bracis-33630	221	32	20,000	20,000	NUM
bracis-33630	221	33	.	.	PUNCT
bracis-33630	222	1	then	then	ADV
bracis-33630	222	2	,	,	PUNCT
bracis-33630	222	3	it	it	PRON
bracis-33630	222	4	reduces	reduce	VERB
bracis-33630	222	5	to	to	ADP
bracis-33630	222	6	0.100	0.100	NUM
bracis-33630	222	7	for	for	ADP
bracis-33630	222	8	a	a	DET
bracis-33630	222	9	dataset	dataset	NOUN
bracis-33630	222	10	size	size	NOUN
bracis-33630	222	11	of	of	ADP
bracis-33630	222	12	30,000	30,000	NUM
bracis-33630	222	13	.	.	PUNCT
bracis-33630	223	1	\(\beta	\(\beta	PROPN
bracis-33630	223	2	\	\	NOUN
bracis-33630	223	3	)	)	PUNCT
bracis-33630	223	4	values	value	NOUN
bracis-33630	223	5	are	be	AUX
bracis-33630	223	6	significantly	significantly	ADV
bracis-33630	223	7	lower	low	ADJ
bracis-33630	223	8	for	for	ADP
bracis-33630	223	9	the	the	DET
bracis-33630	223	10	best	good	ADJ
bracis-33630	223	11	nmi	nmi	PROPN
bracis-33630	223	12	cases	case	NOUN
bracis-33630	223	13	,	,	PUNCT
bracis-33630	223	14	starting	start	VERB
bracis-33630	223	15	at	at	ADP
bracis-33630	223	16	0.006	0.006	NUM
bracis-33630	223	17	and	and	CCONJ
bracis-33630	223	18	0.004	0.004	NUM
bracis-33630	223	19	for	for	ADP
bracis-33630	223	20	dataset	dataset	ADJ
bracis-33630	223	21	sizes	size	NOUN
bracis-33630	223	22	of	of	ADP
bracis-33630	223	23	10,000	10,000	NUM
bracis-33630	223	24	and	and	CCONJ
bracis-33630	223	25	20,000	20,000	NUM
bracis-33630	223	26	,	,	PUNCT
bracis-33630	223	27	respectively	respectively	ADV
bracis-33630	223	28	.	.	PUNCT
bracis-33630	224	1	this	this	DET
bracis-33630	224	2	trend	trend	NOUN
bracis-33630	224	3	continues	continue	VERB
bracis-33630	224	4	as	as	ADP
bracis-33630	224	5	the	the	DET
bracis-33630	224	6	dataset	dataset	NOUN
bracis-33630	224	7	size	size	NOUN
bracis-33630	224	8	increases	increase	NOUN
bracis-33630	224	9	,	,	PUNCT
bracis-33630	224	10	reflecting	reflect	VERB
bracis-33630	224	11	the	the	DET
bracis-33630	224	12	parameter	parameter	NOUN
bracis-33630	224	13	’s	’s	PART
bracis-33630	224	14	role	role	NOUN
bracis-33630	224	15	in	in	ADP
bracis-33630	224	16	optimizing	optimize	VERB
bracis-33630	224	17	cluster	cluster	NOUN
bracis-33630	224	18	quality	quality	NOUN
bracis-33630	224	19	.	.	PUNCT
bracis-33630	225	1	the	the	DET
bracis-33630	225	2	reduction	reduction	NOUN
bracis-33630	225	3	in	in	ADP
bracis-33630	225	4	\(\beta	\(\beta	PRON
bracis-33630	225	5	\	\	NOUN
bracis-33630	225	6	)	)	PUNCT
bracis-33630	225	7	values	value	NOUN
bracis-33630	225	8	for	for	ADP
bracis-33630	225	9	the	the	DET
bracis-33630	225	10	best	good	ADJ
bracis-33630	225	11	nmi	nmi	NOUN
bracis-33630	225	12	case	case	NOUN
bracis-33630	225	13	suggests	suggest	VERB
bracis-33630	225	14	a	a	DET
bracis-33630	225	15	focus	focus	NOUN
bracis-33630	225	16	on	on	ADP
bracis-33630	225	17	enhancing	enhance	VERB
bracis-33630	225	18	the	the	DET
bracis-33630	225	19	granularity	granularity	NOUN
bracis-33630	225	20	of	of	ADP
bracis-33630	225	21	clustering	clustering	ADJ
bracis-33630	225	22	results	result	NOUN
bracis-33630	225	23	to	to	PART
bracis-33630	225	24	achieve	achieve	VERB
bracis-33630	225	25	a	a	DET
bracis-33630	225	26	higher	high	ADJ
bracis-33630	225	27	normalized	normalize	VERB
bracis-33630	225	28	mutual	mutual	ADJ
bracis-33630	225	29	information	information	NOUN
bracis-33630	225	30	score	score	NOUN
bracis-33630	225	31	.	.	PUNCT
bracis-33630	226	1	6	6	NUM
bracis-33630	226	2	conclusions	conclusion	NOUN
bracis-33630	226	3	our	our	PRON
bracis-33630	226	4	work	work	NOUN
bracis-33630	226	5	demonstrates	demonstrate	VERB
bracis-33630	226	6	that	that	SCONJ
bracis-33630	226	7	lda	lda	PROPN
bracis-33630	226	8	-	-	PUNCT
bracis-33630	226	9	based	base	VERB
bracis-33630	226	10	methods	method	NOUN
bracis-33630	226	11	face	face	VERB
bracis-33630	226	12	significant	significant	ADJ
bracis-33630	226	13	difficulties	difficulty	NOUN
bracis-33630	226	14	when	when	SCONJ
bracis-33630	226	15	applied	apply	VERB
bracis-33630	226	16	to	to	ADP
bracis-33630	226	17	large	large	ADJ
bracis-33630	226	18	datasets	dataset	NOUN
bracis-33630	226	19	.	.	PUNCT
bracis-33630	227	1	this	this	PRON
bracis-33630	227	2	is	be	AUX
bracis-33630	227	3	evidenced	evidence	VERB
bracis-33630	227	4	by	by	ADP
bracis-33630	227	5	the	the	DET
bracis-33630	227	6	increasing	increase	VERB
bracis-33630	227	7	processing	processing	NOUN
bracis-33630	227	8	time	time	NOUN
bracis-33630	227	9	proportional	proportional	ADJ
bracis-33630	227	10	to	to	ADP
bracis-33630	227	11	the	the	DET
bracis-33630	227	12	dataset	dataset	NOUN
bracis-33630	227	13	size	size	NOUN
bracis-33630	227	14	,	,	PUNCT
bracis-33630	227	15	highlighting	highlight	VERB
bracis-33630	227	16	the	the	DET
bracis-33630	227	17	scalability	scalability	NOUN
bracis-33630	227	18	challenges	challenge	VERB
bracis-33630	227	19	inherent	inherent	ADJ
bracis-33630	227	20	in	in	ADP
bracis-33630	227	21	traditional	traditional	ADJ
bracis-33630	227	22	lda	lda	PROPN
bracis-33630	227	23	approaches	approach	NOUN
bracis-33630	227	24	.	.	PUNCT
bracis-33630	228	1	it	it	PRON
bracis-33630	228	2	was	be	AUX
bracis-33630	228	3	confirmed	confirm	VERB
bracis-33630	228	4	that	that	SCONJ
bracis-33630	228	5	high	high	ADJ
bracis-33630	228	6	nmi	nmi	NOUN
bracis-33630	228	7	and	and	CCONJ
bracis-33630	228	8	homogeneity	homogeneity	NOUN
bracis-33630	228	9	scores	score	NOUN
bracis-33630	228	10	are	be	AUX
bracis-33630	228	11	often	often	ADV
bracis-33630	228	12	associated	associate	VERB
bracis-33630	228	13	with	with	ADP
bracis-33630	228	14	forming	form	VERB
bracis-33630	228	15	single	single	ADJ
bracis-33630	228	16	-	-	PUNCT
bracis-33630	228	17	element	element	NOUN
bracis-33630	228	18	or	or	CCONJ
bracis-33630	228	19	small	small	ADJ
bracis-33630	228	20	clusters	cluster	NOUN
bracis-33630	228	21	.	.	PUNCT
bracis-33630	229	1	this	this	PRON
bracis-33630	229	2	is	be	AUX
bracis-33630	229	3	evident	evident	ADJ
bracis-33630	229	4	from	from	ADP
bracis-33630	229	5	the	the	DET
bracis-33630	229	6	elevated	elevated	ADJ
bracis-33630	229	7	predict	predict	ADJ
bracis-33630	229	8	/	/	SYM
bracis-33630	229	9	true	true	ADJ
bracis-33630	229	10	size	size	NOUN
bracis-33630	229	11	rate	rate	NOUN
bracis-33630	229	12	observed	observe	VERB
bracis-33630	229	13	.	.	PUNCT
bracis-33630	230	1	however	however	ADV
bracis-33630	230	2	,	,	PUNCT
bracis-33630	230	3	an	an	DET
bracis-33630	230	4	exception	exception	NOUN
bracis-33630	230	5	was	be	AUX
bracis-33630	230	6	noted	note	VERB
bracis-33630	230	7	in	in	ADP
bracis-33630	230	8	the	the	DET
bracis-33630	230	9	gtin	gtin	NOUN
bracis-33630	230	10	dataset	dataset	VERB
bracis-33630	230	11	,	,	PUNCT
bracis-33630	230	12	where	where	SCONJ
bracis-33630	230	13	methods	method	NOUN
bracis-33630	230	14	such	such	ADJ
bracis-33630	230	15	as	as	ADP
bracis-33630	230	16	osdm	osdm	NOUN
bracis-33630	230	17	,	,	PUNCT
bracis-33630	230	18	eindm	eindm	NOUN
bracis-33630	230	19	,	,	PUNCT
bracis-33630	230	20	and	and	CCONJ
bracis-33630	230	21	estream	estream	NOUN
bracis-33630	230	22	did	do	AUX
bracis-33630	230	23	not	not	PART
bracis-33630	230	24	conform	conform	VERB
bracis-33630	230	25	to	to	ADP
bracis-33630	230	26	this	this	DET
bracis-33630	230	27	premise	premise	NOUN
bracis-33630	230	28	,	,	PUNCT
bracis-33630	230	29	indicating	indicate	VERB
bracis-33630	230	30	some	some	DET
bracis-33630	230	31	variability	variability	NOUN
bracis-33630	230	32	in	in	ADP
bracis-33630	230	33	performance	performance	NOUN
bracis-33630	230	34	across	across	ADP
bracis-33630	230	35	different	different	ADJ
bracis-33630	230	36	datasets	dataset	NOUN
bracis-33630	230	37	.	.	PUNCT
bracis-33630	231	1	evaluating	evaluate	VERB
bracis-33630	231	2	clustering	clustering	NOUN
bracis-33630	231	3	based	base	VERB
bracis-33630	231	4	on	on	ADP
bracis-33630	231	5	the	the	DET
bracis-33630	231	6	best	good	ADJ
bracis-33630	231	7	aic	aic	PROPN
bracis-33630	231	8	demonstrated	demonstrate	VERB
bracis-33630	231	9	effectiveness	effectiveness	NOUN
bracis-33630	231	10	in	in	ADP
bracis-33630	231	11	controlling	control	VERB
bracis-33630	231	12	single	single	ADJ
bracis-33630	231	13	-	-	PUNCT
bracis-33630	231	14	element	element	NOUN
bracis-33630	231	15	clusters	cluster	NOUN
bracis-33630	231	16	.	.	PUNCT
bracis-33630	232	1	however	however	ADV
bracis-33630	232	2	,	,	PUNCT
bracis-33630	232	3	this	this	PRON
bracis-33630	232	4	came	come	VERB
bracis-33630	232	5	at	at	ADP
bracis-33630	232	6	the	the	DET
bracis-33630	232	7	cost	cost	NOUN
bracis-33630	232	8	of	of	ADP
bracis-33630	232	9	reduced	reduced	ADJ
bracis-33630	232	10	nmi	nmi	PROPN
bracis-33630	232	11	,	,	PUNCT
bracis-33630	232	12	homogeneity	homogeneity	NOUN
bracis-33630	232	13	,	,	PUNCT
bracis-33630	232	14	purity	purity	NOUN
bracis-33630	232	15	,	,	PUNCT
bracis-33630	232	16	and	and	CCONJ
bracis-33630	232	17	completeness	completeness	NOUN
bracis-33630	232	18	.	.	PUNCT
bracis-33630	233	1	this	this	DET
bracis-33630	233	2	trade	trade	NOUN
bracis-33630	233	3	-	-	PUNCT
bracis-33630	233	4	off	off	NOUN
bracis-33630	233	5	suggests	suggest	VERB
bracis-33630	233	6	that	that	SCONJ
bracis-33630	233	7	while	while	SCONJ
bracis-33630	233	8	aic	aic	PROPN
bracis-33630	233	9	can	can	AUX
bracis-33630	233	10	mitigate	mitigate	VERB
bracis-33630	233	11	overly	overly	ADV
bracis-33630	233	12	simplistic	simplistic	ADJ
bracis-33630	233	13	clustering	clustering	ADJ
bracis-33630	233	14	solutions	solution	NOUN
bracis-33630	233	15	,	,	PUNCT
bracis-33630	233	16	it	it	PRON
bracis-33630	233	17	may	may	AUX
bracis-33630	233	18	also	also	ADV
bracis-33630	233	19	impact	impact	VERB
bracis-33630	233	20	the	the	DET
bracis-33630	233	21	overall	overall	ADJ
bracis-33630	233	22	quality	quality	NOUN
bracis-33630	233	23	of	of	ADP
bracis-33630	233	24	the	the	DET
bracis-33630	233	25	clustering	clustering	ADJ
bracis-33630	233	26	results	result	NOUN
bracis-33630	233	27	.	.	PUNCT
bracis-33630	234	1	notes	note	VERB
bracis-33630	234	2	1.the	1.the	DET
bracis-33630	234	3	southern	southern	ADJ
bracis-33630	234	4	common	common	ADJ
bracis-33630	234	5	market	market	NOUN
bracis-33630	234	6	(	(	PUNCT
bracis-33630	234	7	mercosur	mercosur	PROPN
bracis-33630	234	8	for	for	ADP
bracis-33630	234	9	its	its	PRON
bracis-33630	234	10	spanish	spanish	ADJ
bracis-33630	234	11	initials	initial	NOUN
bracis-33630	234	12	)	)	PUNCT
bracis-33630	234	13	is	be	AUX
bracis-33630	234	14	a	a	DET
bracis-33630	234	15	regional	regional	ADJ
bracis-33630	234	16	integration	integration	NOUN
bracis-33630	234	17	process	process	NOUN
bracis-33630	234	18	,	,	PUNCT
bracis-33630	234	19	initially	initially	ADV
bracis-33630	234	20	established	establish	VERB
bracis-33630	234	21	by	by	ADP
bracis-33630	234	22	argentina	argentina	PROPN
bracis-33630	234	23	,	,	PUNCT
bracis-33630	234	24	brazil	brazil	PROPN
bracis-33630	234	25	,	,	PUNCT
bracis-33630	234	26	paraguay	paraguay	NOUN
bracis-33630	234	27	and	and	CCONJ
bracis-33630	234	28	uruguay	uruguay	NOUN
bracis-33630	234	29	,	,	PUNCT
bracis-33630	234	30	and	and	CCONJ
bracis-33630	234	31	subsequently	subsequently	ADV
bracis-33630	234	32	joined	join	VERB
bracis-33630	234	33	by	by	ADP
bracis-33630	234	34	venezuela	venezuela	PROPN
bracis-33630	234	35	and	and	CCONJ
bracis-33630	234	36	bolivia	bolivia	PROPN
bracis-33630	234	37	*	*	PROPN
bracis-33630	234	38	.	.	PUNCT
bracis-33630	235	1	2.https://github.com/jaykumarr/osdm	2.https://github.com/jaykumarr/osdm	X
bracis-33630	235	2	.	.	PUNCT
bracis-33630	236	1	3.https://github.com/jaykumarr/osgm	3.https://github.com/jaykumarr/osgm	NUM
bracis-33630	236	2	.	.	PUNCT
bracis-33630	237	1	4.https://github.com/jaykumarr/eindm	4.https://github.com/jaykumarr/eindm	NOUN
bracis-33630	237	2	.	.	PUNCT
bracis-33630	238	1	5.https://github.com/rashadulrakib/short-text-stream-clustering/tree/master/onlineclustering	5.https://github.com/rashadulrakib/short-text-stream-clustering/tree/master/onlineclustering	NOUN
bracis-33630	238	2	.	.	PUNCT
bracis-33630	239	1	references	reference	NOUN
bracis-33630	239	2	cheng	cheng	PROPN
bracis-33630	239	3	,	,	PUNCT
bracis-33630	239	4	x.	x.	PROPN
bracis-33630	239	5	,	,	PUNCT
bracis-33630	239	6	yan	yan	PROPN
bracis-33630	239	7	,	,	PUNCT
bracis-33630	239	8	x.	x.	PROPN
bracis-33630	239	9	,	,	PUNCT
bracis-33630	239	10	lan	lan	PROPN
bracis-33630	239	11	,	,	PUNCT
bracis-33630	239	12	y.	y.	PROPN
bracis-33630	239	13	,	,	PUNCT
bracis-33630	239	14	guo	guo	PROPN
bracis-33630	239	15	,	,	PUNCT
bracis-33630	239	16	j.	j.	PROPN
bracis-33630	239	17	:	:	PROPN
bracis-33630	239	18	btm	btm	PROPN
bracis-33630	239	19	:	:	PUNCT
bracis-33630	239	20	topic	topic	NOUN
bracis-33630	239	21	modeling	modeling	NOUN
bracis-33630	239	22	over	over	ADP
bracis-33630	239	23	short	short	ADJ
bracis-33630	239	24	texts	text	NOUN
bracis-33630	239	25	.	.	PUNCT
bracis-33630	240	1	in	in	ADP
bracis-33630	240	2	:	:	PUNCT
bracis-33630	240	3	ieee	ieee	NOUN
bracis-33630	240	4	transactions	transaction	NOUN
bracis-33630	240	5	on	on	ADP
bracis-33630	240	6	knowledge	knowledge	NOUN
bracis-33630	240	7	and	and	CCONJ
bracis-33630	240	8	data	datum	NOUN
bracis-33630	240	9	engineering	engineering	NOUN
bracis-33630	240	10	,	,	PUNCT
bracis-33630	240	11	vol	vol	NOUN
bracis-33630	240	12	.	.	PROPN
bracis-33630	240	13	26	26	NUM
bracis-33630	240	14	,	,	PUNCT
bracis-33630	240	15	no	no	INTJ
bracis-33630	240	16	.	.	NOUN
bracis-33630	240	17	12	12	NUM
bracis-33630	240	18	,	,	PUNCT
bracis-33630	240	19	pp	pp	ADJ
bracis-33630	240	20	.	.	PUNCT
bracis-33630	241	1	2928–2941	2928–2941	NUM
bracis-33630	241	2	(	(	PUNCT
bracis-33630	241	3	2014	2014	NUM
bracis-33630	241	4	)	)	PUNCT
bracis-33630	241	5	.	.	PUNCT
bracis-33630	242	1	https://doi.org/10.1109/tkde.2014.2313872	https://doi.org/10.1109/tkde.2014.2313872	PROPN
bracis-33630	242	2	zhang	zhang	PROPN
bracis-33630	242	3	,	,	PUNCT
bracis-33630	242	4	p.	p.	NOUN
bracis-33630	242	5	:	:	PUNCT
bracis-33630	242	6	evaluating	evaluate	VERB
bracis-33630	242	7	accuracy	accuracy	NOUN
bracis-33630	242	8	of	of	ADP
bracis-33630	242	9	community	community	NOUN
bracis-33630	242	10	detection	detection	NOUN
bracis-33630	242	11	using	use	VERB
bracis-33630	242	12	the	the	DET
bracis-33630	242	13	relative	relative	ADJ
bracis-33630	242	14	normalized	normalize	VERB
bracis-33630	242	15	mutual	mutual	ADJ
bracis-33630	242	16	information	information	NOUN
bracis-33630	242	17	.	.	PUNCT
bracis-33630	243	1	j.	j.	PROPN
bracis-33630	243	2	stat	stat	PROPN
bracis-33630	243	3	.	.	PUNCT
bracis-33630	244	1	mech	mech	PROPN
bracis-33630	244	2	:	:	PUNCT
bracis-33630	244	3	theory	theory	NOUN
bracis-33630	244	4	exp	exp	NOUN
bracis-33630	244	5	.	.	PROPN
bracis-33630	244	6	2015(11	2015(11	NUM
bracis-33630	244	7	)	)	PUNCT
bracis-33630	244	8	,	,	PUNCT
bracis-33630	244	9	p11006	p11006	NUM
bracis-33630	244	10	(	(	PUNCT
bracis-33630	244	11	2015	2015	NUM
bracis-33630	244	12	)	)	PUNCT
bracis-33630	244	13	article	article	NOUN
bracis-33630	244	14	  	  	SPACE
bracis-33630	244	15	mathscinet	mathscinet	NOUN
bracis-33630	244	16	  	  	SPACE
bracis-33630	244	17	math	math	NOUN
bracis-33630	244	18	  	  	SPACE
bracis-33630	244	19	google	google	PROPN
bracis-33630	244	20	scholar	scholar	NOUN
bracis-33630	244	21	  	  	SPACE
bracis-33630	244	22	jerdee	jerdee	ADJ
bracis-33630	244	23	,	,	PUNCT
bracis-33630	244	24	m.	m.	NOUN
bracis-33630	244	25	,	,	PUNCT
bracis-33630	244	26	kirkley	kirkley	PROPN
bracis-33630	244	27	,	,	PUNCT
bracis-33630	244	28	a.	a.	PROPN
bracis-33630	244	29	,	,	PUNCT
bracis-33630	244	30	newman	newman	PROPN
bracis-33630	244	31	,	,	PUNCT
bracis-33630	244	32	m.e.j	m.e.j	PROPN
bracis-33630	244	33	.	.	PUNCT
bracis-33630	244	34	:	:	PUNCT
bracis-33630	244	35	normalized	normalize	VERB
bracis-33630	244	36	mutual	mutual	ADJ
bracis-33630	244	37	information	information	NOUN
bracis-33630	244	38	is	be	AUX
bracis-33630	244	39	a	a	DET
bracis-33630	244	40	biased	biased	ADJ
bracis-33630	244	41	measure	measure	NOUN
bracis-33630	244	42	for	for	ADP
bracis-33630	244	43	classification	classification	NOUN
bracis-33630	244	44	and	and	CCONJ
bracis-33630	244	45	community	community	NOUN
bracis-33630	244	46	detection	detection	NOUN
bracis-33630	244	47	.	.	PUNCT
bracis-33630	245	1	arxiv	arxiv	PROPN
bracis-33630	245	2	preprint	preprint	VERB
bracis-33630	245	3	arxiv:2307.01282	arxiv:2307.01282	NOUN
bracis-33630	245	4	(	(	PUNCT
bracis-33630	245	5	2023	2023	NUM
bracis-33630	245	6	)	)	PUNCT
bracis-33630	245	7	devlin	devlin	NOUN
bracis-33630	245	8	,	,	PUNCT
bracis-33630	245	9	j.	j.	PROPN
bracis-33630	245	10	,	,	PUNCT
bracis-33630	245	11	chang	chang	PROPN
bracis-33630	245	12	,	,	PUNCT
bracis-33630	245	13	m.-w	m.-w	PROPN
bracis-33630	245	14	.	.	PROPN
bracis-33630	245	15	,	,	PUNCT
bracis-33630	245	16	lee	lee	PROPN
bracis-33630	245	17	,	,	PUNCT
bracis-33630	245	18	k.	k.	PROPN
bracis-33630	245	19	,	,	PUNCT
bracis-33630	245	20	toutanova	toutanova	PROPN
bracis-33630	245	21	,	,	PUNCT
bracis-33630	245	22	k.	k.	PROPN
bracis-33630	245	23	:	:	PUNCT
bracis-33630	246	1	bert	bert	PROPN
bracis-33630	246	2	:	:	PUNCT
bracis-33630	246	3	pre	pre	ADJ
bracis-33630	246	4	-	-	NOUN
bracis-33630	246	5	training	training	NOUN
bracis-33630	246	6	of	of	ADP
bracis-33630	246	7	deep	deep	ADJ
bracis-33630	246	8	bidirectional	bidirectional	ADJ
bracis-33630	246	9	transformers	transformer	NOUN
bracis-33630	246	10	for	for	ADP
bracis-33630	246	11	language	language	NOUN
bracis-33630	246	12	understanding	understanding	NOUN
bracis-33630	246	13	.	.	PUNCT
bracis-33630	247	1	arxiv	arxiv	PROPN
bracis-33630	247	2	preprint	preprint	PROPN
bracis-33630	247	3	arxiv:1810.04805	arxiv:1810.04805	PROPN
bracis-33630	247	4	(	(	PUNCT
bracis-33630	247	5	2018	2018	NUM
bracis-33630	247	6	)	)	PUNCT
bracis-33630	247	7	campello	campello	PROPN
bracis-33630	247	8	,	,	PUNCT
bracis-33630	247	9	r.j.g.b	r.j.g.b	PROPN
bracis-33630	247	10	.	.	PROPN
bracis-33630	247	11	,	,	PUNCT
bracis-33630	247	12	moulavi	moulavi	PROPN
bracis-33630	247	13	,	,	PUNCT
bracis-33630	247	14	d.	d.	PROPN
bracis-33630	247	15	,	,	PUNCT
bracis-33630	247	16	sander	sander	PROPN
bracis-33630	247	17	,	,	PUNCT
bracis-33630	247	18	j.	j.	PROPN
bracis-33630	247	19	:	:	PUNCT
bracis-33630	247	20	density	density	NOUN
bracis-33630	247	21	-	-	PUNCT
bracis-33630	247	22	based	base	VERB
bracis-33630	247	23	clustering	clustering	NOUN
bracis-33630	247	24	based	base	VERB
bracis-33630	247	25	on	on	ADP
bracis-33630	247	26	hierarchical	hierarchical	ADJ
bracis-33630	247	27	density	density	NOUN
bracis-33630	247	28	estimates	estimate	NOUN
bracis-33630	247	29	.	.	PUNCT
bracis-33630	248	1	in	in	ADP
bracis-33630	248	2	:	:	PUNCT
bracis-33630	248	3	pacific	pacific	ADJ
bracis-33630	248	4	-	-	PUNCT
bracis-33630	248	5	asia	asia	PROPN
bracis-33630	248	6	conference	conference	NOUN
bracis-33630	248	7	on	on	ADP
bracis-33630	248	8	knowledge	knowledge	NOUN
bracis-33630	248	9	discovery	discovery	PROPN
bracis-33630	248	10	and	and	CCONJ
bracis-33630	248	11	data	datum	NOUN
bracis-33630	248	12	mining	mining	NOUN
bracis-33630	248	13	,	,	PUNCT
bracis-33630	248	14	pp	pp	ADJ
bracis-33630	248	15	.	.	PUNCT
bracis-33630	249	1	160–172	160–172	NUM
bracis-33630	249	2	.	.	PUNCT
bracis-33630	250	1	springer	springer	NOUN
bracis-33630	250	2	(	(	PUNCT
bracis-33630	250	3	2013	2013	NUM
bracis-33630	250	4	)	)	PUNCT
bracis-33630	250	5	google	google	PROPN
bracis-33630	250	6	scholar	scholar	NOUN
bracis-33630	250	7	  	  	SPACE
bracis-33630	250	8	dai	dai	PROPN
bracis-33630	250	9	,	,	PUNCT
bracis-33630	250	10	z.	z.	PROPN
bracis-33630	250	11	,	,	PUNCT
bracis-33630	250	12	callan	callan	PROPN
bracis-33630	250	13	,	,	PUNCT
bracis-33630	250	14	j.	j.	PROPN
bracis-33630	250	15	:	:	PUNCT
bracis-33630	250	16	deeper	deep	ADJ
bracis-33630	250	17	text	text	NOUN
bracis-33630	250	18	understanding	understanding	NOUN
bracis-33630	250	19	for	for	ADP
bracis-33630	250	20	ir	ir	PROPN
bracis-33630	250	21	with	with	ADP
bracis-33630	250	22	contextual	contextual	ADJ
bracis-33630	250	23	neural	neural	ADJ
bracis-33630	250	24	language	language	NOUN
bracis-33630	250	25	modeling	modeling	NOUN
bracis-33630	250	26	.	.	PUNCT
bracis-33630	251	1	in	in	ADP
bracis-33630	251	2	:	:	PUNCT
bracis-33630	251	3	proceedings	proceeding	NOUN
bracis-33630	251	4	of	of	ADP
bracis-33630	251	5	the	the	DET
bracis-33630	251	6	42nd	42nd	PROPN
bracis-33630	251	7	international	international	PROPN
bracis-33630	251	8	acm	acm	PROPN
bracis-33630	251	9	sigir	sigir	PROPN
bracis-33630	251	10	conference	conference	PROPN
bracis-33630	251	11	on	on	ADP
bracis-33630	251	12	research	research	NOUN
bracis-33630	251	13	and	and	CCONJ
bracis-33630	251	14	development	development	NOUN
bracis-33630	251	15	in	in	ADP
bracis-33630	251	16	information	information	NOUN
bracis-33630	251	17	retrieval	retrieval	NOUN
bracis-33630	251	18	(	(	PUNCT
bracis-33630	251	19	2019	2019	NUM
bracis-33630	251	20	)	)	PUNCT
bracis-33630	251	21	google	google	PROPN
bracis-33630	251	22	scholar	scholar	NOUN
bracis-33630	251	23	  	  	SPACE
bracis-33630	251	24	asyaky	asyaky	PROPN
bracis-33630	251	25	,	,	PUNCT
bracis-33630	251	26	m.s	m.s	PROPN
bracis-33630	251	27	.	.	PROPN
bracis-33630	251	28	,	,	PUNCT
bracis-33630	251	29	mandala	mandala	PROPN
bracis-33630	251	30	,	,	PUNCT
bracis-33630	251	31	r.	r.	PROPN
bracis-33630	251	32	:	:	PUNCT
bracis-33630	251	33	improving	improve	VERB
bracis-33630	251	34	the	the	DET
bracis-33630	251	35	performance	performance	NOUN
bracis-33630	251	36	of	of	ADP
bracis-33630	251	37	hdbscan	hdbscan	NOUN
bracis-33630	251	38	on	on	ADP
bracis-33630	251	39	short	short	ADJ
bracis-33630	251	40	text	text	NOUN
bracis-33630	251	41	clustering	cluster	VERB
bracis-33630	251	42	by	by	ADP
bracis-33630	251	43	using	use	VERB
bracis-33630	251	44	word	word	NOUN
bracis-33630	251	45	embedding	embed	VERB
bracis-33630	251	46	and	and	CCONJ
bracis-33630	251	47	umap	umap	ADJ
bracis-33630	251	48	.	.	PUNCT
bracis-33630	252	1	in	in	ADP
bracis-33630	252	2	:	:	PUNCT
bracis-33630	252	3	8th	8th	ADJ
bracis-33630	252	4	international	international	ADJ
bracis-33630	252	5	conference	conference	NOUN
bracis-33630	252	6	on	on	ADP
bracis-33630	252	7	advanced	advanced	ADJ
bracis-33630	252	8	informatics	informatic	NOUN
bracis-33630	252	9	:	:	PUNCT
bracis-33630	252	10	concepts	concept	NOUN
bracis-33630	252	11	,	,	PUNCT
bracis-33630	252	12	theory	theory	NOUN
bracis-33630	252	13	,	,	PUNCT
bracis-33630	252	14	and	and	CCONJ
bracis-33630	252	15	applications	application	NOUN
bracis-33630	252	16	(	(	PUNCT
bracis-33630	252	17	icaicta	icaicta	NOUN
bracis-33630	252	18	)	)	PUNCT
bracis-33630	252	19	,	,	PUNCT
bracis-33630	252	20	pp	pp	ADV
bracis-33630	252	21	.	.	PUNCT
bracis-33630	253	1	1–6	1–6	X
bracis-33630	253	2	.	.	X
bracis-33630	254	1	ieee	ieee	NOUN
bracis-33630	254	2	(	(	PUNCT
bracis-33630	254	3	2021	2021	NUM
bracis-33630	254	4	)	)	PUNCT
bracis-33630	254	5	google	google	PROPN
bracis-33630	254	6	scholar	scholar	NOUN
bracis-33630	254	7	  	  	SPACE
bracis-33630	254	8	eklund	eklund	PROPN
bracis-33630	254	9	,	,	PUNCT
bracis-33630	254	10	a.	a.	NOUN
bracis-33630	254	11	,	,	PUNCT
bracis-33630	254	12	forsman	forsman	NOUN
bracis-33630	254	13	,	,	PUNCT
bracis-33630	254	14	m.	m.	NOUN
bracis-33630	254	15	:	:	PUNCT
bracis-33630	254	16	topic	topic	NOUN
bracis-33630	254	17	modeling	modeling	NOUN
bracis-33630	254	18	by	by	ADP
bracis-33630	254	19	clustering	cluster	VERB
bracis-33630	254	20	language	language	NOUN
bracis-33630	254	21	model	model	NOUN
bracis-33630	254	22	embeddings	embedding	NOUN
bracis-33630	254	23	:	:	PUNCT
bracis-33630	254	24	human	human	ADJ
bracis-33630	254	25	validation	validation	NOUN
bracis-33630	254	26	on	on	ADP
bracis-33630	254	27	an	an	DET
bracis-33630	254	28	industry	industry	NOUN
bracis-33630	254	29	dataset	dataset	NOUN
bracis-33630	254	30	.	.	PUNCT
bracis-33630	255	1	in	in	ADP
bracis-33630	255	2	:	:	PUNCT
bracis-33630	255	3	proceedings	proceeding	NOUN
bracis-33630	255	4	of	of	ADP
bracis-33630	255	5	the	the	DET
bracis-33630	255	6	2022	2022	NUM
bracis-33630	255	7	conference	conference	NOUN
bracis-33630	255	8	on	on	ADP
bracis-33630	255	9	empirical	empirical	ADJ
bracis-33630	255	10	methods	method	NOUN
bracis-33630	255	11	in	in	ADP
bracis-33630	255	12	natural	natural	ADJ
bracis-33630	255	13	language	language	NOUN
bracis-33630	255	14	processing	processing	NOUN
bracis-33630	255	15	:	:	PUNCT
bracis-33630	255	16	industry	industry	NOUN
bracis-33630	255	17	track	track	NOUN
bracis-33630	255	18	,	,	PUNCT
bracis-33630	255	19	pp	pp	ADJ
bracis-33630	255	20	.	.	PUNCT
bracis-33630	256	1	635–643	635–643	NUM
bracis-33630	256	2	(	(	PUNCT
bracis-33630	256	3	2022	2022	NUM
bracis-33630	256	4	)	)	PUNCT
bracis-33630	256	5	google	google	PROPN
bracis-33630	256	6	scholar	scholar	NOUN
bracis-33630	256	7	  	  	SPACE
bracis-33630	256	8	schulte	schulte	PROPN
bracis-33630	256	9	,	,	PUNCT
bracis-33630	256	10	j.p	j.p	PROPN
bracis-33630	256	11	.	.	PROPN
bracis-33630	256	12	,	,	PUNCT
bracis-33630	256	13	et	et	PROPN
bracis-33630	256	14	al	al	PROPN
bracis-33630	256	15	.	.	PROPN
bracis-33630	256	16	:	:	PUNCT
bracis-33630	257	1	elinac	elinac	ADJ
bracis-33630	257	2	:	:	PUNCT
bracis-33630	257	3	autoencoder	autoencoder	NOUN
bracis-33630	257	4	approach	approach	NOUN
bracis-33630	257	5	for	for	ADP
bracis-33630	257	6	electronic	electronic	ADJ
bracis-33630	257	7	invoices	invoice	NOUN
bracis-33630	257	8	data	datum	NOUN
bracis-33630	257	9	clustering	cluster	VERB
bracis-33630	257	10	.	.	PUNCT
bracis-33630	258	1	appl	appl	PROPN
bracis-33630	258	2	.	.	PUNCT
bracis-33630	259	1	sci	sci	PROPN
bracis-33630	259	2	.	.	PROPN
bracis-33630	259	3	12(6	12(6	NUM
bracis-33630	259	4	)	)	PUNCT
bracis-33630	259	5	,	,	PUNCT
bracis-33630	259	6	3008	3008	NUM
bracis-33630	259	7	(	(	PUNCT
bracis-33630	259	8	2022	2022	NUM
bracis-33630	259	9	)	)	PUNCT
bracis-33630	259	10	google	google	PROPN
bracis-33630	259	11	scholar	scholar	NOUN
bracis-33630	259	12	  	  	SPACE
bracis-33630	259	13	andrade	andrade	PROPN
bracis-33630	259	14	,	,	PUNCT
bracis-33630	259	15	c.	c.	PROPN
bracis-33630	259	16	,	,	PUNCT
bracis-33630	259	17	ribeiro	ribeiro	PROPN
bracis-33630	259	18	,	,	PUNCT
bracis-33630	259	19	r.p	r.p	PROPN
bracis-33630	259	20	.	.	PROPN
bracis-33630	259	21	,	,	PUNCT
bracis-33630	259	22	gama	gama	PROPN
bracis-33630	259	23	,	,	PUNCT
bracis-33630	259	24	j.	j.	PROPN
bracis-33630	259	25	:	:	PUNCT
bracis-33630	259	26	topic	topic	NOUN
bracis-33630	259	27	model	model	NOUN
bracis-33630	259	28	with	with	ADP
bracis-33630	259	29	contextual	contextual	ADJ
bracis-33630	259	30	outlier	outlier	NOUN
bracis-33630	259	31	handling	handling	NOUN
bracis-33630	259	32	:	:	PUNCT
bracis-33630	259	33	a	a	DET
bracis-33630	259	34	study	study	NOUN
bracis-33630	259	35	on	on	ADP
bracis-33630	259	36	electronic	electronic	ADJ
bracis-33630	259	37	invoice	invoice	NOUN
bracis-33630	259	38	product	product	NOUN
bracis-33630	259	39	descriptions	description	NOUN
bracis-33630	259	40	.	.	PUNCT
bracis-33630	260	1	in	in	ADP
bracis-33630	260	2	:	:	PUNCT
bracis-33630	260	3	epia	epia	ADJ
bracis-33630	260	4	conference	conference	NOUN
bracis-33630	260	5	on	on	ADP
bracis-33630	260	6	artificial	artificial	ADJ
bracis-33630	260	7	intelligence	intelligence	NOUN
bracis-33630	260	8	,	,	PUNCT
bracis-33630	260	9	pp	pp	ADJ
bracis-33630	260	10	.	.	PUNCT
bracis-33630	261	1	365–377	365–377	NUM
bracis-33630	261	2	.	.	PUNCT
bracis-33630	261	3	springer	springer	NOUN
bracis-33630	261	4	(	(	PUNCT
bracis-33630	261	5	2023	2023	NUM
bracis-33630	261	6	)	)	PUNCT
bracis-33630	261	7	google	google	NOUN
bracis-33630	261	8	scholar	scholar	NOUN
bracis-33630	261	9	  	  	SPACE
bracis-33630	261	10	blei	blei	NOUN
bracis-33630	261	11	,	,	PUNCT
bracis-33630	261	12	d.m	d.m	PROPN
bracis-33630	261	13	.	.	PROPN
bracis-33630	261	14	,	,	PUNCT
bracis-33630	261	15	ng	ng	PROPN
bracis-33630	261	16	,	,	PUNCT
bracis-33630	261	17	a.y	a.y	PROPN
bracis-33630	261	18	.	.	PROPN
bracis-33630	261	19	,	,	PUNCT
bracis-33630	261	20	jordan	jordan	PROPN
bracis-33630	261	21	,	,	PUNCT
bracis-33630	261	22	m.i	m.i	PROPN
bracis-33630	261	23	.	.	PROPN
bracis-33630	261	24	:	:	PUNCT
bracis-33630	262	1	latent	latent	PROPN
bracis-33630	262	2	dirichlet	dirichlet	PROPN
bracis-33630	262	3	allocation	allocation	NOUN
bracis-33630	262	4	.	.	PUNCT
bracis-33630	263	1	j.	j.	PROPN
bracis-33630	263	2	mach	mach	PROPN
bracis-33630	263	3	.	.	PUNCT
bracis-33630	264	1	learn	learn	VERB
bracis-33630	264	2	.	.	PUNCT
bracis-33630	265	1	res	re	NOUN
bracis-33630	265	2	.	.	PUNCT
bracis-33630	266	1	3(jan	3(jan	NUM
bracis-33630	266	2	)	)	PUNCT
bracis-33630	266	3	,	,	PUNCT
bracis-33630	266	4	993–1022	993–1022	NUM
bracis-33630	266	5	(	(	PUNCT
bracis-33630	266	6	2003	2003	NUM
bracis-33630	266	7	)	)	PUNCT
bracis-33630	266	8	google	google	PROPN
bracis-33630	266	9	scholar	scholar	NOUN
bracis-33630	266	10	  	  	SPACE
bracis-33630	266	11	liang	liang	PROPN
bracis-33630	266	12	,	,	PUNCT
bracis-33630	266	13	s.	s.	PROPN
bracis-33630	266	14	,	,	PUNCT
bracis-33630	266	15	yilmaz	yilmaz	PROPN
bracis-33630	266	16	,	,	PUNCT
bracis-33630	266	17	e.	e.	PROPN
bracis-33630	266	18	,	,	PUNCT
bracis-33630	266	19	kanoulas	kanoulas	PROPN
bracis-33630	266	20	,	,	PUNCT
bracis-33630	266	21	e.	e.	PROPN
bracis-33630	266	22	:	:	PUNCT
bracis-33630	266	23	dynamic	dynamic	ADJ
bracis-33630	266	24	clustering	clustering	NOUN
bracis-33630	266	25	of	of	ADP
bracis-33630	266	26	streaming	stream	VERB
bracis-33630	266	27	short	short	ADJ
bracis-33630	266	28	documents	document	NOUN
bracis-33630	266	29	.	.	PUNCT
bracis-33630	267	1	in	in	ADP
bracis-33630	267	2	:	:	PUNCT
bracis-33630	267	3	proceedings	proceeding	NOUN
bracis-33630	267	4	of	of	ADP
bracis-33630	267	5	the	the	DET
bracis-33630	267	6	22nd	22nd	PROPN
bracis-33630	267	7	acm	acm	PROPN
bracis-33630	267	8	sigkdd	sigkdd	PROPN
bracis-33630	267	9	international	international	ADJ
bracis-33630	267	10	conference	conference	NOUN
bracis-33630	267	11	on	on	ADP
bracis-33630	267	12	knowledge	knowledge	NOUN
bracis-33630	267	13	discovery	discovery	PROPN
bracis-33630	267	14	and	and	CCONJ
bracis-33630	267	15	data	datum	NOUN
bracis-33630	267	16	mining	mining	NOUN
bracis-33630	267	17	,	,	PUNCT
bracis-33630	267	18	pp	pp	ADP
bracis-33630	267	19	.	.	PUNCT
bracis-33630	268	1	995–1004	995–1004	NUM
bracis-33630	268	2	(	(	PUNCT
bracis-33630	268	3	2016	2016	NUM
bracis-33630	268	4	)	)	PUNCT
bracis-33630	268	5	google	google	PROPN
bracis-33630	268	6	scholar	scholar	NOUN
bracis-33630	268	7	  	  	SPACE
bracis-33630	268	8	yin	yin	PROPN
bracis-33630	268	9	,	,	PUNCT
bracis-33630	268	10	j.	j.	PROPN
bracis-33630	268	11	,	,	PUNCT
bracis-33630	268	12	wang	wang	PROPN
bracis-33630	268	13	,	,	PUNCT
bracis-33630	268	14	j.	j.	PROPN
bracis-33630	268	15	,	,	PUNCT
bracis-33630	268	16	xu	xu	PROPN
bracis-33630	268	17	,	,	PUNCT
bracis-33630	268	18	w.	w.	PROPN
bracis-33630	268	19	,	,	PUNCT
bracis-33630	268	20	gao	gao	PROPN
bracis-33630	268	21	,	,	PUNCT
bracis-33630	268	22	m.	m.	NOUN
bracis-33630	268	23	:	:	PUNCT
bracis-33630	268	24	model	model	NOUN
bracis-33630	268	25	-	-	PUNCT
bracis-33630	268	26	based	base	VERB
bracis-33630	268	27	clustering	clustering	NOUN
bracis-33630	268	28	of	of	ADP
bracis-33630	268	29	short	short	ADJ
bracis-33630	268	30	text	text	NOUN
bracis-33630	268	31	streams	stream	NOUN
bracis-33630	268	32	.	.	PUNCT
bracis-33630	269	1	in	in	ADP
bracis-33630	269	2	:	:	PUNCT
bracis-33630	269	3	27th	27th	ADJ
bracis-33630	269	4	acm	acm	PROPN
bracis-33630	269	5	international	international	ADJ
bracis-33630	269	6	conference	conference	NOUN
bracis-33630	269	7	on	on	ADP
bracis-33630	269	8	information	information	NOUN
bracis-33630	269	9	and	and	CCONJ
bracis-33630	269	10	knowledge	knowledge	NOUN
bracis-33630	269	11	management	management	NOUN
bracis-33630	269	12	,	,	PUNCT
bracis-33630	269	13	pp	pp	ADJ
bracis-33630	269	14	.	.	PUNCT
bracis-33630	270	1	697–706	697–706	NUM
bracis-33630	270	2	.	.	PUNCT
bracis-33630	270	3	acm	acm	PROPN
bracis-33630	270	4	(	(	PUNCT
bracis-33630	270	5	2018	2018	NUM
bracis-33630	270	6	)	)	PUNCT
bracis-33630	270	7	google	google	PROPN
bracis-33630	270	8	scholar	scholar	NOUN
bracis-33630	270	9	  	  	SPACE
bracis-33630	270	10	chen	chen	PROPN
bracis-33630	270	11	,	,	PUNCT
bracis-33630	270	12	j.	j.	PROPN
bracis-33630	270	13	,	,	PUNCT
bracis-33630	270	14	gong	gong	PROPN
bracis-33630	270	15	,	,	PUNCT
bracis-33630	270	16	z.	z.	PROPN
bracis-33630	270	17	,	,	PUNCT
bracis-33630	270	18	liu	liu	PROPN
bracis-33630	270	19	,	,	PUNCT
bracis-33630	270	20	w.	w.	PROPN
bracis-33630	270	21	:	:	PUNCT
bracis-33630	270	22	a	a	DET
bracis-33630	270	23	nonparametric	nonparametric	NOUN
bracis-33630	270	24	model	model	NOUN
bracis-33630	270	25	for	for	ADP
bracis-33630	270	26	online	online	ADJ
bracis-33630	270	27	topic	topic	NOUN
bracis-33630	270	28	discovery	discovery	NOUN
bracis-33630	270	29	with	with	ADP
bracis-33630	270	30	word	word	NOUN
bracis-33630	270	31	embeddings	embedding	NOUN
bracis-33630	270	32	.	.	PUNCT
bracis-33630	271	1	inf	inf	PROPN
bracis-33630	271	2	.	.	PUNCT
bracis-33630	272	1	sci	sci	PROPN
bracis-33630	272	2	.	.	PROPN
bracis-33630	272	3	504	504	NUM
bracis-33630	272	4	,	,	PUNCT
bracis-33630	272	5	32–47	32–47	NUM
bracis-33630	272	6	(	(	PUNCT
bracis-33630	272	7	2019	2019	NUM
bracis-33630	272	8	)	)	PUNCT
bracis-33630	272	9	.	.	PUNCT
bracis-33630	273	1	elsevier	elsevier	PROPN
bracis-33630	273	2	google	google	PROPN
bracis-33630	273	3	scholar	scholar	PROPN
bracis-33630	273	4	  	  	SPACE
bracis-33630	273	5	chen	chen	PROPN
bracis-33630	273	6	,	,	PUNCT
bracis-33630	273	7	j.	j.	PROPN
bracis-33630	273	8	,	,	PUNCT
bracis-33630	273	9	gong	gong	PROPN
bracis-33630	273	10	,	,	PUNCT
bracis-33630	273	11	z.	z.	PROPN
bracis-33630	273	12	,	,	PUNCT
bracis-33630	273	13	liu	liu	PROPN
bracis-33630	273	14	,	,	PUNCT
bracis-33630	273	15	w.	w.	PROPN
bracis-33630	273	16	:	:	PUNCT
bracis-33630	273	17	a	a	DET
bracis-33630	273	18	dirichlet	dirichlet	NOUN
bracis-33630	273	19	process	process	NOUN
bracis-33630	273	20	biterm	biterm	NOUN
bracis-33630	273	21	-	-	PUNCT
bracis-33630	273	22	based	base	VERB
bracis-33630	273	23	mixture	mixture	NOUN
bracis-33630	273	24	model	model	NOUN
bracis-33630	273	25	for	for	ADP
bracis-33630	273	26	short	short	ADJ
bracis-33630	273	27	text	text	NOUN
bracis-33630	273	28	stream	stream	NOUN
bracis-33630	273	29	clustering	clustering	NOUN
bracis-33630	273	30	.	.	PUNCT
bracis-33630	274	1	appl	appl	PROPN
bracis-33630	274	2	.	.	PUNCT
bracis-33630	275	1	intell	intell	PROPN
bracis-33630	275	2	.	.	PUNCT
bracis-33630	276	1	50	50	NUM
bracis-33630	276	2	,	,	PUNCT
bracis-33630	276	3	1609–1619	1609–1619	NUM
bracis-33630	276	4	(	(	PUNCT
bracis-33630	276	5	2020	2020	NUM
bracis-33630	276	6	)	)	PUNCT
bracis-33630	276	7	.	.	PUNCT
bracis-33630	277	1	springer	springer	NOUN
bracis-33630	277	2	google	google	PROPN
bracis-33630	277	3	scholar	scholar	PROPN
bracis-33630	277	4	  	  	SPACE
bracis-33630	277	5	xu	xu	PROPN
bracis-33630	277	6	,	,	PUNCT
bracis-33630	277	7	y.	y.	PROPN
bracis-33630	277	8	,	,	PUNCT
bracis-33630	277	9	wang	wang	PROPN
bracis-33630	277	10	,	,	PUNCT
bracis-33630	277	11	s.	s.	PROPN
bracis-33630	277	12	,	,	PUNCT
bracis-33630	277	13	zhang	zhang	PROPN
bracis-33630	277	14	,	,	PUNCT
bracis-33630	277	15	s.	s.	PROPN
bracis-33630	277	16	,	,	PUNCT
bracis-33630	277	17	wang	wang	PROPN
bracis-33630	277	18	,	,	PUNCT
bracis-33630	277	19	f.	f.	PROPN
bracis-33630	277	20	:	:	PUNCT
bracis-33630	277	21	dynamic	dynamic	ADJ
bracis-33630	277	22	clustering	clustering	NOUN
bracis-33630	277	23	for	for	ADP
bracis-33630	277	24	short	short	ADJ
bracis-33630	277	25	text	text	NOUN
bracis-33630	277	26	stream	stream	NOUN
bracis-33630	277	27	based	base	VERB
bracis-33630	277	28	on	on	ADP
bracis-33630	277	29	dirichlet	dirichlet	PROPN
bracis-33630	277	30	process	process	NOUN
bracis-33630	277	31	.	.	PUNCT
bracis-33630	278	1	ieee	ieee	NOUN
bracis-33630	278	2	access	access	NOUN
bracis-33630	278	3	10	10	NUM
bracis-33630	278	4	,	,	PUNCT
bracis-33630	278	5	22852–22865	22852–22865	NUM
bracis-33630	278	6	(	(	PUNCT
bracis-33630	278	7	2022	2022	NUM
bracis-33630	278	8	)	)	PUNCT
bracis-33630	278	9	math	math	NOUN
bracis-33630	278	10	  	  	SPACE
bracis-33630	278	11	google	google	PROPN
bracis-33630	278	12	scholar	scholar	NOUN
bracis-33630	278	13	  	  	SPACE
bracis-33630	278	14	rakib	rakib	NOUN
bracis-33630	278	15	,	,	PUNCT
bracis-33630	278	16	m.r.h	m.r.h	PROPN
bracis-33630	278	17	.	.	PROPN
bracis-33630	278	18	,	,	PUNCT
bracis-33630	278	19	asaduzzaman	asaduzzaman	NOUN
bracis-33630	278	20	,	,	PUNCT
bracis-33630	278	21	m.	m.	NOUN
bracis-33630	278	22	:	:	PUNCT
bracis-33630	278	23	fast	fast	ADJ
bracis-33630	278	24	clustering	clustering	NOUN
bracis-33630	278	25	of	of	ADP
bracis-33630	278	26	short	short	ADJ
bracis-33630	278	27	text	text	NOUN
bracis-33630	278	28	streams	stream	NOUN
bracis-33630	278	29	using	use	VERB
bracis-33630	278	30	efficient	efficient	ADJ
bracis-33630	278	31	cluster	cluster	NOUN
bracis-33630	278	32	indexing	indexing	NOUN
bracis-33630	278	33	and	and	CCONJ
bracis-33630	278	34	dynamic	dynamic	ADJ
bracis-33630	278	35	similarity	similarity	NOUN
bracis-33630	278	36	thresholds	threshold	NOUN
bracis-33630	278	37	.	.	PUNCT
bracis-33630	279	1	corr	corr	NOUN
bracis-33630	279	2	abs/2101.08595	abs/2101.08595	NOUN
bracis-33630	279	3	(	(	PUNCT
bracis-33630	279	4	2021	2021	NUM
bracis-33630	279	5	)	)	PUNCT
bracis-33630	279	6	google	google	NOUN
bracis-33630	279	7	scholar	scholar	NOUN
bracis-33630	279	8	  	  	SPACE
bracis-33630	279	9	rakib	rakib	NOUN
bracis-33630	279	10	,	,	PUNCT
bracis-33630	279	11	m.r.h	m.r.h	PROPN
bracis-33630	279	12	.	.	PROPN
bracis-33630	279	13	,	,	PUNCT
bracis-33630	279	14	zeh	zeh	PROPN
bracis-33630	279	15	,	,	PUNCT
bracis-33630	279	16	n.	n.	NOUN
bracis-33630	279	17	,	,	PUNCT
bracis-33630	279	18	jankowska	jankowska	PROPN
bracis-33630	279	19	,	,	PUNCT
bracis-33630	279	20	m.	m.	NOUN
bracis-33630	279	21	,	,	PUNCT
bracis-33630	279	22	milios	milio	NOUN
bracis-33630	279	23	,	,	PUNCT
bracis-33630	279	24	e.	e.	PROPN
bracis-33630	279	25	:	:	PUNCT
bracis-33630	279	26	enhancement	enhancement	NOUN
bracis-33630	279	27	of	of	ADP
bracis-33630	279	28	short	short	ADJ
bracis-33630	279	29	text	text	NOUN
bracis-33630	279	30	clustering	cluster	VERB
bracis-33630	279	31	by	by	ADP
bracis-33630	279	32	iterative	iterative	ADJ
bracis-33630	279	33	classification	classification	NOUN
bracis-33630	279	34	.	.	PUNCT
bracis-33630	280	1	in	in	ADP
bracis-33630	280	2	:	:	PUNCT
bracis-33630	280	3	natural	natural	ADJ
bracis-33630	280	4	language	language	NOUN
bracis-33630	280	5	processing	processing	NOUN
bracis-33630	280	6	and	and	CCONJ
bracis-33630	280	7	information	information	NOUN
bracis-33630	280	8	systems	system	NOUN
bracis-33630	280	9	:	:	PUNCT
bracis-33630	280	10	25th	25th	ADJ
bracis-33630	280	11	international	international	ADJ
bracis-33630	280	12	conference	conference	NOUN
bracis-33630	280	13	on	on	ADP
bracis-33630	280	14	applications	application	NOUN
bracis-33630	280	15	of	of	ADP
bracis-33630	280	16	natural	natural	ADJ
bracis-33630	280	17	language	language	NOUN
bracis-33630	280	18	to	to	ADP
bracis-33630	280	19	information	information	NOUN
bracis-33630	280	20	systems	system	NOUN
bracis-33630	280	21	,	,	PUNCT
bracis-33630	280	22	nldb	nldb	ADJ
bracis-33630	280	23	2020	2020	NUM
bracis-33630	280	24	,	,	PUNCT
bracis-33630	280	25	saarbrücken	saarbrücken	PROPN
bracis-33630	280	26	,	,	PUNCT
bracis-33630	280	27	germany	germany	PROPN
bracis-33630	280	28	,	,	PUNCT
bracis-33630	280	29	june	june	PROPN
bracis-33630	280	30	24–26	24–26	NUM
bracis-33630	280	31	,	,	PUNCT
bracis-33630	280	32	2020	2020	NUM
bracis-33630	280	33	,	,	PUNCT
bracis-33630	280	34	proceedings	proceeding	NOUN
bracis-33630	280	35	25	25	NUM
bracis-33630	280	36	,	,	PUNCT
bracis-33630	280	37	pp	pp	ADJ
bracis-33630	280	38	.	.	PUNCT
bracis-33630	281	1	105–117	105–117	NUM
bracis-33630	281	2	.	.	PUNCT
bracis-33630	281	3	springer	springer	NOUN
bracis-33630	281	4	(	(	PUNCT
bracis-33630	281	5	2020	2020	NUM
bracis-33630	281	6	)	)	PUNCT
bracis-33630	281	7	google	google	PROPN
bracis-33630	281	8	scholar	scholar	NOUN
bracis-33630	281	9	  	  	SPACE
bracis-33630	281	10	kumar	kumar	PROPN
bracis-33630	281	11	,	,	PUNCT
bracis-33630	281	12	j.	j.	PROPN
bracis-33630	281	13	,	,	PUNCT
bracis-33630	281	14	shao	shao	PROPN
bracis-33630	281	15	,	,	PUNCT
bracis-33630	281	16	j.	j.	PROPN
bracis-33630	281	17	,	,	PUNCT
bracis-33630	281	18	uddin	uddin	PROPN
bracis-33630	281	19	,	,	PUNCT
bracis-33630	281	20	s.	s.	PROPN
bracis-33630	281	21	,	,	PUNCT
bracis-33630	281	22	ali	ali	PROPN
bracis-33630	281	23	,	,	PUNCT
bracis-33630	281	24	w.	w.	PROPN
bracis-33630	281	25	:	:	PUNCT
bracis-33630	281	26	an	an	DET
bracis-33630	281	27	online	online	ADJ
bracis-33630	281	28	semantic	semantic	ADV
bracis-33630	281	29	-	-	PUNCT
bracis-33630	281	30	enhanced	enhance	VERB
bracis-33630	281	31	dirichlet	dirichlet	PROPN
bracis-33630	281	32	model	model	NOUN
bracis-33630	281	33	for	for	ADP
bracis-33630	281	34	short	short	ADJ
bracis-33630	281	35	text	text	NOUN
bracis-33630	281	36	stream	stream	NOUN
bracis-33630	281	37	clustering	cluster	VERB
bracis-33630	281	38	.	.	PUNCT
bracis-33630	282	1	in	in	ADP
bracis-33630	282	2	:	:	PUNCT
bracis-33630	282	3	proceedings	proceeding	NOUN
bracis-33630	282	4	of	of	ADP
bracis-33630	282	5	the	the	DET
bracis-33630	282	6	58th	58th	ADJ
bracis-33630	282	7	annual	annual	ADJ
bracis-33630	282	8	meeting	meeting	NOUN
bracis-33630	282	9	of	of	ADP
bracis-33630	282	10	the	the	DET
bracis-33630	282	11	association	association	NOUN
bracis-33630	282	12	for	for	ADP
bracis-33630	282	13	computational	computational	ADJ
bracis-33630	282	14	linguistics	linguistic	NOUN
bracis-33630	282	15	,	,	PUNCT
bracis-33630	282	16	pp	pp	ADJ
bracis-33630	282	17	.	.	PUNCT
bracis-33630	283	1	766–776	766–776	NUM
bracis-33630	283	2	.	.	PUNCT
bracis-33630	283	3	association	association	NOUN
bracis-33630	283	4	for	for	ADP
bracis-33630	283	5	computational	computational	ADJ
bracis-33630	283	6	linguistics	linguistic	NOUN
bracis-33630	283	7	(	(	PUNCT
bracis-33630	283	8	2020	2020	NUM
bracis-33630	283	9	)	)	PUNCT
bracis-33630	283	10	google	google	PROPN
bracis-33630	283	11	scholar	scholar	NOUN
bracis-33630	283	12	  	  	SPACE
bracis-33630	283	13	kumar	kumar	PROPN
bracis-33630	283	14	,	,	PUNCT
bracis-33630	283	15	j.	j.	PROPN
bracis-33630	283	16	,	,	PUNCT
bracis-33630	283	17	ud	ud	INTJ
bracis-33630	283	18	din	din	PROPN
bracis-33630	283	19	,	,	PUNCT
bracis-33630	283	20	s.	s.	PROPN
bracis-33630	283	21	,	,	PUNCT
bracis-33630	283	22	yang	yang	PROPN
bracis-33630	283	23	,	,	PUNCT
bracis-33630	283	24	q.	q.	PROPN
bracis-33630	283	25	,	,	PUNCT
bracis-33630	283	26	kumar	kumar	PROPN
bracis-33630	283	27	,	,	PUNCT
bracis-33630	283	28	r.	r.	PROPN
bracis-33630	283	29	,	,	PUNCT
bracis-33630	283	30	shao	shao	PROPN
bracis-33630	283	31	,	,	PUNCT
bracis-33630	283	32	j.	j.	PROPN
bracis-33630	283	33	:	:	PUNCT
bracis-33630	283	34	an	an	DET
bracis-33630	283	35	online	online	ADJ
bracis-33630	283	36	semantic	semantic	ADV
bracis-33630	283	37	-	-	PUNCT
bracis-33630	283	38	enhanced	enhance	VERB
bracis-33630	283	39	graphical	graphical	ADJ
bracis-33630	283	40	model	model	NOUN
bracis-33630	283	41	for	for	ADP
bracis-33630	283	42	evolving	evolve	VERB
bracis-33630	283	43	short	short	ADJ
bracis-33630	283	44	text	text	NOUN
bracis-33630	283	45	stream	stream	NOUN
bracis-33630	283	46	clustering	clustering	NOUN
bracis-33630	283	47	.	.	PUNCT
bracis-33630	284	1	ieee	ieee	PROPN
bracis-33630	284	2	trans	trans	PROPN
bracis-33630	284	3	.	.	PUNCT
bracis-33630	285	1	cybern	cybern	PROPN
bracis-33630	285	2	.	.	PUNCT
bracis-33630	286	1	52(12	52(12	NUM
bracis-33630	286	2	)	)	PUNCT
bracis-33630	286	3	,	,	PUNCT
bracis-33630	286	4	13809–13820	13809–13820	NUM
bracis-33630	286	5	(	(	PUNCT
bracis-33630	286	6	2021	2021	NUM
bracis-33630	286	7	)	)	PUNCT
bracis-33630	286	8	google	google	PROPN
bracis-33630	286	9	scholar	scholar	NOUN
bracis-33630	286	10	  	  	SPACE
bracis-33630	286	11	kumar	kumar	PROPN
bracis-33630	286	12	,	,	PUNCT
bracis-33630	286	13	j.	j.	PROPN
bracis-33630	286	14	,	,	PUNCT
bracis-33630	286	15	shao	shao	PROPN
bracis-33630	286	16	,	,	PUNCT
bracis-33630	286	17	j.	j.	PROPN
bracis-33630	286	18	,	,	PUNCT
bracis-33630	286	19	kumar	kumar	PROPN
bracis-33630	286	20	,	,	PUNCT
bracis-33630	286	21	r.	r.	PROPN
bracis-33630	286	22	,	,	PUNCT
bracis-33630	286	23	ud	ud	INTJ
bracis-33630	286	24	din	din	PROPN
bracis-33630	286	25	,	,	PUNCT
bracis-33630	286	26	s.	s.	PROPN
bracis-33630	286	27	,	,	PUNCT
bracis-33630	286	28	mawuli	mawuli	PROPN
bracis-33630	286	29	,	,	PUNCT
bracis-33630	286	30	c.b	c.b	PROPN
bracis-33630	286	31	.	.	PROPN
bracis-33630	286	32	,	,	PUNCT
bracis-33630	286	33	yang	yang	PROPN
bracis-33630	286	34	,	,	PUNCT
bracis-33630	286	35	q.	q.	PROPN
bracis-33630	286	36	:	:	PUNCT
bracis-33630	286	37	a	a	DET
bracis-33630	286	38	context	context	NOUN
bracis-33630	286	39	-	-	PUNCT
bracis-33630	286	40	enhanced	enhance	VERB
bracis-33630	286	41	dirichlet	dirichlet	NOUN
bracis-33630	286	42	model	model	NOUN
bracis-33630	286	43	for	for	ADP
bracis-33630	286	44	online	online	ADJ
bracis-33630	286	45	clustering	clustering	NOUN
bracis-33630	286	46	in	in	ADP
bracis-33630	286	47	short	short	ADJ
bracis-33630	286	48	text	text	NOUN
bracis-33630	286	49	streams	stream	NOUN
bracis-33630	286	50	.	.	PUNCT
bracis-33630	287	1	in	in	ADP
bracis-33630	287	2	:	:	PUNCT
bracis-33630	287	3	expert	expert	NOUN
bracis-33630	287	4	systems	system	NOUN
bracis-33630	287	5	with	with	ADP
bracis-33630	287	6	applications	application	NOUN
bracis-33630	287	7	,	,	PUNCT
bracis-33630	287	8	vol	vol	NOUN
bracis-33630	287	9	.	.	PROPN
bracis-33630	287	10	228	228	NUM
bracis-33630	287	11	,	,	PUNCT
bracis-33630	287	12	p.	p.	NOUN
bracis-33630	287	13	120262	120262	NUM
bracis-33630	287	14	(	(	PUNCT
bracis-33630	287	15	2023	2023	NUM
bracis-33630	287	16	)	)	PUNCT
bracis-33630	287	17	google	google	NOUN
bracis-33630	287	18	scholar	scholar	NOUN
bracis-33630	287	19	  	  	SPACE
bracis-33630	287	20	rakib	rakib	NOUN
bracis-33630	287	21	,	,	PUNCT
bracis-33630	287	22	m.r.h	m.r.h	PROPN
bracis-33630	287	23	.	.	PROPN
bracis-33630	287	24	,	,	PUNCT
bracis-33630	287	25	zeh	zeh	PROPN
bracis-33630	287	26	,	,	PUNCT
bracis-33630	287	27	n.	n.	NOUN
bracis-33630	287	28	,	,	PUNCT
bracis-33630	287	29	milios	milio	NOUN
bracis-33630	287	30	,	,	PUNCT
bracis-33630	287	31	e.	e.	PROPN
bracis-33630	287	32	:	:	PUNCT
bracis-33630	287	33	efficient	efficient	ADJ
bracis-33630	287	34	clustering	clustering	NOUN
bracis-33630	287	35	of	of	ADP
bracis-33630	287	36	short	short	ADJ
bracis-33630	287	37	text	text	NOUN
bracis-33630	287	38	streams	stream	NOUN
bracis-33630	287	39	using	use	VERB
bracis-33630	287	40	online	online	ADJ
bracis-33630	287	41	-	-	PUNCT
bracis-33630	287	42	offline	offline	ADJ
bracis-33630	287	43	clustering	clustering	NOUN
bracis-33630	287	44	.	.	PUNCT
bracis-33630	288	1	in	in	ADP
bracis-33630	288	2	:	:	PUNCT
bracis-33630	288	3	proceedings	proceeding	NOUN
bracis-33630	288	4	of	of	ADP
bracis-33630	288	5	the	the	DET
bracis-33630	288	6	21st	21st	ADJ
bracis-33630	288	7	acm	acm	NOUN
bracis-33630	288	8	symposium	symposium	NOUN
bracis-33630	288	9	on	on	ADP
bracis-33630	288	10	document	document	NOUN
bracis-33630	288	11	engineering	engineering	NOUN
bracis-33630	288	12	,	,	PUNCT
bracis-33630	288	13	pp	pp	ADP
bracis-33630	288	14	.	.	PUNCT
bracis-33630	289	1	1–10	1–10	NOUN
bracis-33630	289	2	(	(	PUNCT
bracis-33630	289	3	2021	2021	NUM
bracis-33630	289	4	)	)	PUNCT
bracis-33630	289	5	google	google	NOUN
bracis-33630	289	6	scholar	scholar	NOUN
bracis-33630	289	7	  	  	SPACE
bracis-33630	289	8	murshed	murshed	ADJ
bracis-33630	289	9	,	,	PUNCT
bracis-33630	289	10	b.a.h	b.a.h	PROPN
bracis-33630	289	11	.	.	PROPN
bracis-33630	289	12	,	,	PUNCT
bracis-33630	289	13	et	et	PROPN
bracis-33630	289	14	al	al	PROPN
bracis-33630	289	15	.	.	PUNCT
bracis-33630	289	16	:	:	PUNCT
bracis-33630	290	1	short	short	ADJ
bracis-33630	290	2	text	text	NOUN
bracis-33630	290	3	topic	topic	NOUN
bracis-33630	290	4	modelling	model	VERB
bracis-33630	290	5	approaches	approach	NOUN
bracis-33630	290	6	in	in	ADP
bracis-33630	290	7	the	the	DET
bracis-33630	290	8	context	context	NOUN
bracis-33630	290	9	of	of	ADP
bracis-33630	290	10	big	big	ADJ
bracis-33630	290	11	data	datum	NOUN
bracis-33630	290	12	:	:	PUNCT
bracis-33630	290	13	taxonomy	taxonomy	NOUN
bracis-33630	290	14	,	,	PUNCT
bracis-33630	290	15	survey	survey	NOUN
bracis-33630	290	16	,	,	PUNCT
bracis-33630	290	17	and	and	CCONJ
bracis-33630	290	18	analysis	analysis	NOUN
bracis-33630	290	19	.	.	PUNCT
bracis-33630	291	1	artif	artif	INTJ
bracis-33630	291	2	.	.	PUNCT
bracis-33630	292	1	intell	intell	PROPN
bracis-33630	292	2	.	.	PUNCT
bracis-33630	293	1	rev	rev	PROPN
bracis-33630	293	2	.	.	PROPN
bracis-33630	294	1	56(6	56(6	NUM
bracis-33630	294	2	)	)	PUNCT
bracis-33630	295	1	,	,	PUNCT
bracis-33630	295	2	5133–5260	5133–5260	NUM
bracis-33630	295	3	(	(	PUNCT
bracis-33630	295	4	2023	2023	NUM
bracis-33630	295	5	)	)	PUNCT
bracis-33630	295	6	google	google	NOUN
bracis-33630	295	7	scholar	scholar	NOUN
bracis-33630	295	8	  	  	SPACE
bracis-33630	295	9	download	download	NOUN
bracis-33630	295	10	references	reference	NOUN
bracis-33630	295	11	acknowledgments	acknowledgment	NOUN
bracis-33630	295	12	the	the	DET
bracis-33630	295	13	authors	author	NOUN
bracis-33630	295	14	wish	wish	VERB
bracis-33630	295	15	to	to	PART
bracis-33630	295	16	clarify	clarify	VERB
bracis-33630	295	17	that	that	SCONJ
bracis-33630	295	18	the	the	DET
bracis-33630	295	19	first	first	ADJ
bracis-33630	295	20	author	author	NOUN
bracis-33630	295	21	received	receive	VERB
bracis-33630	295	22	support	support	NOUN
bracis-33630	295	23	from	from	ADP
bracis-33630	295	24	amazonas	amazonas	PROPN
bracis-33630	295	25	state	state	NOUN
bracis-33630	295	26	government	government	PROPN
bracis-33630	295	27	/	/	SYM
bracis-33630	295	28	brazil	brazil	PROPN
bracis-33630	295	29	for	for	ADP
bracis-33630	295	30	this	this	DET
bracis-33630	295	31	research	research	NOUN
bracis-33630	295	32	project	project	NOUN
bracis-33630	295	33	.	.	PUNCT
bracis-33630	296	1	author	author	NOUN
bracis-33630	296	2	information	information	NOUN
bracis-33630	296	3	authors	author	NOUN
bracis-33630	296	4	and	and	CCONJ
bracis-33630	296	5	affiliations	affiliation	NOUN
bracis-33630	296	6	department	department	PROPN
bracis-33630	296	7	of	of	ADP
bracis-33630	296	8	computer	computer	NOUN
bracis-33630	296	9	science	science	NOUN
bracis-33630	296	10	,	,	PUNCT
bracis-33630	296	11	faculty	faculty	NOUN
bracis-33630	296	12	of	of	ADP
bracis-33630	296	13	sciences	science	NOUN
bracis-33630	296	14	,	,	PUNCT
bracis-33630	296	15	university	university	NOUN
bracis-33630	296	16	of	of	ADP
bracis-33630	296	17	porto	porto	PROPN
bracis-33630	296	18	,	,	PUNCT
bracis-33630	296	19	4169	4169	NUM
bracis-33630	296	20	-	-	SYM
bracis-33630	296	21	007	007	NUM
bracis-33630	296	22	,	,	PUNCT
bracis-33630	296	23	porto	porto	PROPN
bracis-33630	296	24	,	,	PUNCT
bracis-33630	296	25	portugal	portugal	PROPN
bracis-33630	296	26	cesar	cesar	PROPN
bracis-33630	296	27	andrade	andrade	PROPN
bracis-33630	296	28	 	 	SPACE
bracis-33630	296	29	&	&	CCONJ
bracis-33630	296	30	 	 	SPACE
bracis-33630	296	31	rita	rita	PROPN
bracis-33630	297	1	p.	p.	PROPN
bracis-33630	297	2	ribeiro	ribeiro	PROPN
bracis-33630	297	3	faculty	faculty	PROPN
bracis-33630	297	4	of	of	ADP
bracis-33630	297	5	economics	economic	NOUN
bracis-33630	297	6	,	,	PUNCT
bracis-33630	297	7	university	university	NOUN
bracis-33630	297	8	of	of	ADP
bracis-33630	297	9	porto	porto	PROPN
bracis-33630	297	10	,	,	PUNCT
bracis-33630	297	11	4200	4200	NUM
bracis-33630	297	12	-	-	SYM
bracis-33630	297	13	464	464	NUM
bracis-33630	297	14	,	,	PUNCT
bracis-33630	297	15	porto	porto	PROPN
bracis-33630	297	16	,	,	PUNCT
bracis-33630	297	17	portugal	portugal	PROPN
bracis-33630	297	18	joão	joão	PROPN
bracis-33630	297	19	gama	gama	PROPN
bracis-33630	297	20	faculty	faculty	NOUN
bracis-33630	297	21	of	of	ADP
bracis-33630	297	22	computer	computer	NOUN
bracis-33630	297	23	science	science	NOUN
bracis-33630	297	24	,	,	PUNCT
bracis-33630	297	25	university	university	NOUN
bracis-33630	297	26	of	of	ADP
bracis-33630	297	27	porto	porto	PROPN
bracis-33630	297	28	,	,	PUNCT
bracis-33630	297	29	porto	porto	PROPN
bracis-33630	297	30	,	,	PUNCT
bracis-33630	297	31	portugal	portugal	PROPN
bracis-33630	297	32	joão	joão	PROPN
bracis-33630	297	33	gama	gama	PROPN
bracis-33630	297	34	inesc	inesc	PROPN
bracis-33630	297	35	tec	tec	PROPN
bracis-33630	297	36	,	,	PUNCT
bracis-33630	297	37	4200	4200	NUM
bracis-33630	297	38	-	-	SYM
bracis-33630	297	39	465	465	NUM
bracis-33630	297	40	,	,	PUNCT
bracis-33630	297	41	porto	porto	PROPN
bracis-33630	297	42	,	,	PUNCT
bracis-33630	297	43	portugal	portugal	PROPN
bracis-33630	297	44	cesar	cesar	PROPN
bracis-33630	297	45	andrade	andrade	PROPN
bracis-33630	297	46	,	,	PUNCT
bracis-33630	297	47	 	 	SPACE
bracis-33630	297	48	rita	rita	PROPN
bracis-33630	297	49	p.	p.	PROPN
bracis-33630	297	50	ribeiro	ribeiro	PROPN
bracis-33630	297	51	 	 	SPACE
bracis-33630	297	52	&	&	CCONJ
bracis-33630	297	53	 	 	SPACE
bracis-33630	297	54	joão	joão	PROPN
bracis-33630	297	55	gama	gama	PROPN
bracis-33630	297	56	authors	author	NOUN
bracis-33630	297	57	cesar	cesar	VERB
bracis-33630	297	58	andradeview	andradeview	ADJ
bracis-33630	297	59	author	author	NOUN
bracis-33630	297	60	publications	publication	NOUN
bracis-33630	297	61	search	search	NOUN
bracis-33630	297	62	author	author	NOUN
bracis-33630	297	63	on	on	ADP
bracis-33630	297	64	:	:	PUNCT
bracis-33630	297	65	pubmed	pubmed	PROPN
bracis-33630	297	66	 	 	SPACE
bracis-33630	297	67	google	google	PROPN
bracis-33630	297	68	scholar	scholar	NOUN
bracis-33630	297	69	rita	rita	PROPN
bracis-33630	297	70	p.	p.	PROPN
bracis-33630	297	71	ribeiroview	ribeiroview	NOUN
bracis-33630	297	72	author	author	NOUN
bracis-33630	297	73	publications	publication	NOUN
bracis-33630	297	74	search	search	NOUN
bracis-33630	297	75	author	author	NOUN
bracis-33630	297	76	on	on	ADP
bracis-33630	297	77	:	:	PUNCT
bracis-33630	297	78	pubmed	pubmed	PROPN
bracis-33630	297	79	 	 	SPACE
bracis-33630	297	80	google	google	PROPN
bracis-33630	297	81	scholar	scholar	NOUN
bracis-33630	297	82	joão	joão	PROPN
bracis-33630	297	83	gamaview	gamaview	PROPN
bracis-33630	297	84	author	author	NOUN
bracis-33630	297	85	publications	publication	NOUN
bracis-33630	297	86	search	search	NOUN
bracis-33630	297	87	author	author	NOUN
bracis-33630	297	88	on	on	ADP
bracis-33630	297	89	:	:	PUNCT
bracis-33630	297	90	pubmed	pubmed	PROPN
bracis-33630	297	91	 	 	SPACE
bracis-33630	297	92	google	google	PROPN
bracis-33630	297	93	scholar	scholar	NOUN
bracis-33630	297	94	corresponding	correspond	VERB
bracis-33630	297	95	author	author	NOUN
bracis-33630	297	96	correspondence	correspondence	NOUN
bracis-33630	297	97	to	to	ADP
bracis-33630	297	98	cesar	cesar	PROPN
bracis-33630	297	99	andrade	andrade	PROPN
bracis-33630	297	100	.	.	PUNCT
bracis-33630	298	1	editor	editor	NOUN
bracis-33630	298	2	information	information	NOUN
bracis-33630	298	3	editors	editor	NOUN
bracis-33630	298	4	and	and	CCONJ
bracis-33630	298	5	affiliations	affiliation	NOUN
bracis-33630	298	6	universidade	universidade	VERB
bracis-33630	298	7	federal	federal	PROPN
bracis-33630	298	8	fluminense	fluminense	PROPN
bracis-33630	298	9	,	,	PUNCT
bracis-33630	298	10	niterói	niterói	PROPN
bracis-33630	298	11	,	,	PUNCT
bracis-33630	298	12	brazil	brazil	PROPN
bracis-33630	298	13	aline	aline	PROPN
bracis-33630	298	14	paes	paes	PROPN
bracis-33630	298	15	instituto	instituto	PROPN
bracis-33630	298	16	tecnológico	tecnológico	PROPN
bracis-33630	298	17	de	de	PROPN
bracis-33630	298	18	aeronáutica	aeronáutica	PROPN
bracis-33630	298	19	,	,	PUNCT
bracis-33630	298	20	são	são	PROPN
bracis-33630	298	21	josé	josé	PROPN
bracis-33630	298	22	dos	dos	PROPN
bracis-33630	298	23	campos	campos	PROPN
bracis-33630	298	24	,	,	PUNCT
bracis-33630	298	25	brazil	brazil	PROPN
bracis-33630	298	26	filipe	filipe	PROPN
bracis-33630	298	27	a.	a.	PROPN
bracis-33630	298	28	n.	n.	PROPN
bracis-33630	298	29	verri	verri	PROPN
bracis-33630	298	30	rights	right	NOUN
bracis-33630	298	31	and	and	CCONJ
bracis-33630	298	32	permissions	permission	NOUN
bracis-33630	298	33	reprints	reprint	NOUN
bracis-33630	298	34	and	and	CCONJ
bracis-33630	298	35	permissions	permission	VERB
bracis-33630	298	36	copyright	copyright	NOUN
bracis-33630	298	37	information	information	NOUN
bracis-33630	298	38	©	©	ADP
bracis-33630	298	39	2025	2025	NUM
bracis-33630	298	40	the	the	DET
bracis-33630	298	41	author(s	author(s	NOUN
bracis-33630	298	42	)	)	PUNCT
bracis-33630	298	43	,	,	PUNCT
bracis-33630	298	44	under	under	ADP
bracis-33630	298	45	exclusive	exclusive	ADJ
bracis-33630	298	46	license	license	NOUN
bracis-33630	298	47	to	to	ADP
bracis-33630	298	48	springer	springer	NOUN
bracis-33630	298	49	nature	nature	PROPN
bracis-33630	298	50	switzerland	switzerland	PROPN
bracis-33630	298	51	ag	ag	PROPN
bracis-33630	298	52	about	about	ADP
bracis-33630	298	53	this	this	DET
bracis-33630	298	54	paper	paper	NOUN
bracis-33630	298	55	cite	cite	VERB
bracis-33630	298	56	this	this	DET
bracis-33630	298	57	paper	paper	NOUN
bracis-33630	298	58	andrade	andrade	PROPN
bracis-33630	298	59	,	,	PUNCT
bracis-33630	298	60	c.	c.	PROPN
bracis-33630	298	61	,	,	PUNCT
bracis-33630	298	62	ribeiro	ribeiro	PROPN
bracis-33630	298	63	,	,	PUNCT
bracis-33630	298	64	r.p	r.p	PROPN
bracis-33630	298	65	.	.	PROPN
bracis-33630	298	66	,	,	PUNCT
bracis-33630	298	67	gama	gama	PROPN
bracis-33630	298	68	,	,	PUNCT
bracis-33630	298	69	j.	j.	PROPN
bracis-33630	298	70	(	(	PUNCT
bracis-33630	298	71	2025	2025	NUM
bracis-33630	298	72	)	)	PUNCT
bracis-33630	298	73	.	.	PUNCT
bracis-33630	299	1	evaluating	evaluate	VERB
bracis-33630	299	2	short	short	ADJ
bracis-33630	299	3	text	text	NOUN
bracis-33630	299	4	stream	stream	NOUN
bracis-33630	299	5	clustering	cluster	VERB
bracis-33630	299	6	on	on	ADP
bracis-33630	299	7	 	 	SPACE
bracis-33630	299	8	large	large	ADJ
bracis-33630	299	9	e	e	ADJ
bracis-33630	299	10	-	-	NOUN
bracis-33630	299	11	commerce	commerce	NOUN
bracis-33630	299	12	datasets	dataset	NOUN
bracis-33630	299	13	.	.	PUNCT
bracis-33630	300	1	in	in	ADP
bracis-33630	300	2	:	:	PUNCT
bracis-33630	300	3	paes	paes	PROPN
bracis-33630	300	4	,	,	PUNCT
bracis-33630	300	5	a.	a.	PROPN
bracis-33630	300	6	,	,	PUNCT
bracis-33630	300	7	verri	verri	NOUN
bracis-33630	300	8	,	,	PUNCT
bracis-33630	300	9	f.a.n	f.a.n	ADJ
bracis-33630	300	10	.	.	PUNCT
bracis-33630	300	11	(	(	PUNCT
bracis-33630	300	12	eds	ed	NOUN
bracis-33630	300	13	)	)	PUNCT
bracis-33630	300	14	intelligent	intelligent	ADJ
bracis-33630	300	15	systems	system	NOUN
bracis-33630	300	16	.	.	PUNCT
bracis-33630	300	17	bracis	bracis	PROPN
bracis-33630	300	18	2024	2024	NUM
bracis-33630	300	19	.	.	PUNCT
bracis-33630	301	1	lecture	lecture	NOUN
bracis-33630	301	2	notes	note	NOUN
bracis-33630	301	3	in	in	ADP
bracis-33630	301	4	computer	computer	NOUN
bracis-33630	301	5	science	science	NOUN
bracis-33630	301	6	(	(	PUNCT
bracis-33630	301	7	)	)	PUNCT
bracis-33630	301	8	,	,	PUNCT
bracis-33630	301	9	vol	vol	NOUN
bracis-33630	301	10	15414	15414	NUM
bracis-33630	301	11	.	.	PUNCT
bracis-33630	302	1	springer	springer	NOUN
bracis-33630	302	2	,	,	PUNCT
bracis-33630	302	3	cham	cham	PROPN
bracis-33630	302	4	.	.	PUNCT
bracis-33630	303	1	https://doi.org/10.1007/978-3-031-79035-5_17	https://doi.org/10.1007/978-3-031-79035-5_17	PROPN
bracis-33630	303	2	download	download	NOUN
bracis-33630	303	3	citation	citation	NOUN
bracis-33630	303	4	.ris	.ris	PUNCT
bracis-33630	303	5	.enw	.enw	PROPN
bracis-33630	303	6	.bib	.bib	PUNCT
bracis-33630	304	1	doi	doi	PROPN
bracis-33630	304	2	:	:	PUNCT
bracis-33630	304	3	https://doi.org/10.1007/978-3-031-79035-5_17	https://doi.org/10.1007/978-3-031-79035-5_17	PROPN
bracis-33630	304	4	published	publish	VERB
bracis-33630	304	5	:	:	PUNCT
bracis-33630	304	6	30	30	NUM
bracis-33630	304	7	january	january	PROPN
bracis-33630	304	8	2025	2025	NUM
bracis-33630	304	9	publisher	publisher	NOUN
bracis-33630	304	10	name	name	NOUN
bracis-33630	304	11	:	:	PUNCT
bracis-33630	304	12	springer	springer	NOUN
bracis-33630	304	13	,	,	PUNCT
bracis-33630	304	14	cham	cham	PROPN
bracis-33630	304	15	print	print	PROPN
bracis-33630	304	16	isbn	isbn	PROPN
bracis-33630	304	17	:	:	PUNCT
bracis-33630	304	18	978	978	NUM
bracis-33630	304	19	-	-	SYM
bracis-33630	304	20	3	3	NUM
bracis-33630	304	21	-	-	PUNCT
bracis-33630	304	22	031	031	NUM
bracis-33630	304	23	-	-	PUNCT
bracis-33630	304	24	79034	79034	NUM
bracis-33630	304	25	-	-	SYM
bracis-33630	304	26	8	8	NUM
bracis-33630	304	27	online	online	ADJ
bracis-33630	304	28	isbn	isbn	NOUN
bracis-33630	304	29	:	:	PUNCT
bracis-33630	304	30	978	978	NUM
bracis-33630	304	31	-	-	SYM
bracis-33630	304	32	3	3	NUM
bracis-33630	304	33	-	-	PUNCT
bracis-33630	304	34	031	031	NUM
bracis-33630	304	35	-	-	PUNCT
bracis-33630	304	36	79035	79035	NUM
bracis-33630	304	37	-	-	PUNCT
bracis-33630	304	38	5	5	NUM
bracis-33630	304	39	ebook	ebook	NOUN
bracis-33630	304	40	packages	package	NOUN
bracis-33630	304	41	:	:	PUNCT
bracis-33630	304	42	computer	computer	NOUN
bracis-33630	304	43	sciencecomputer	sciencecomputer	NOUN
bracis-33630	304	44	science	science	NOUN
bracis-33630	304	45	(	(	PUNCT
bracis-33630	304	46	r0	r0	NOUN
bracis-33630	304	47	)	)	PUNCT
bracis-33630	304	48	share	share	VERB
bracis-33630	304	49	this	this	DET
bracis-33630	304	50	paper	paper	NOUN
bracis-33630	304	51	anyone	anyone	PRON
bracis-33630	304	52	you	you	PRON
bracis-33630	304	53	share	share	VERB
bracis-33630	304	54	the	the	DET
bracis-33630	304	55	following	follow	VERB
bracis-33630	304	56	link	link	NOUN
bracis-33630	304	57	with	with	ADP
bracis-33630	304	58	will	will	AUX
bracis-33630	304	59	be	be	AUX
bracis-33630	304	60	able	able	ADJ
bracis-33630	304	61	to	to	PART
bracis-33630	304	62	read	read	VERB
bracis-33630	304	63	this	this	DET
bracis-33630	304	64	content	content	NOUN
bracis-33630	304	65	:	:	PUNCT
bracis-33630	304	66	get	get	VERB
bracis-33630	304	67	shareable	shareable	ADJ
bracis-33630	304	68	linksorry	linksorry	NOUN
bracis-33630	304	69	,	,	PUNCT
bracis-33630	304	70	a	a	DET
bracis-33630	304	71	shareable	shareable	ADJ
bracis-33630	304	72	link	link	NOUN
bracis-33630	304	73	is	be	AUX
bracis-33630	304	74	not	not	PART
bracis-33630	304	75	currently	currently	ADV
bracis-33630	304	76	available	available	ADJ
bracis-33630	304	77	for	for	ADP
bracis-33630	304	78	this	this	DET
bracis-33630	304	79	article	article	NOUN
bracis-33630	304	80	.	.	PUNCT
bracis-33630	305	1	copy	copy	VERB
bracis-33630	305	2	shareable	shareable	ADJ
bracis-33630	305	3	link	link	NOUN
bracis-33630	305	4	to	to	PART
bracis-33630	305	5	clipboard	clipboard	NOUN
bracis-33630	305	6	provided	provide	VERB
bracis-33630	305	7	by	by	ADP
bracis-33630	305	8	the	the	DET
bracis-33630	305	9	springer	springer	NOUN
bracis-33630	305	10	nature	nature	PROPN
bracis-33630	305	11	sharedit	sharedit	PROPN
bracis-33630	305	12	content	content	NOUN
bracis-33630	305	13	-	-	PUNCT
bracis-33630	305	14	sharing	share	VERB
bracis-33630	305	15	initiative	initiative	NOUN
bracis-33630	305	16	keywords	keyword	NOUN
bracis-33630	305	17	short	short	ADJ
bracis-33630	305	18	text	text	NOUN
bracis-33630	305	19	stream	stream	NOUN
bracis-33630	305	20	clustering	cluster	VERB
bracis-33630	305	21	latent	latent	ADJ
bracis-33630	305	22	dirichlet	dirichlet	PROPN
bracis-33630	305	23	allocation	allocation	NOUN
bracis-33630	305	24	online	online	ADV
bracis-33630	305	25	publish	publish	VERB
bracis-33630	305	26	with	with	ADP
bracis-33630	305	27	us	us	PROPN
bracis-33630	305	28	policies	policy	NOUN
bracis-33630	305	29	and	and	CCONJ
bracis-33630	305	30	ethics	ethic	NOUN
bracis-33630	305	31	profiles	profile	NOUN
bracis-33630	305	32	cesar	cesar	PROPN
bracis-33630	305	33	andrade	andrade	PROPN
bracis-33630	305	34	view	view	PROPN
bracis-33630	305	35	author	author	NOUN
bracis-33630	305	36	profile	profile	PROPN
bracis-33630	305	37	rita	rita	PROPN
bracis-33630	305	38	p.	p.	PROPN
bracis-33630	305	39	ribeiro	ribeiro	PROPN
bracis-33630	305	40	view	view	PROPN
bracis-33630	305	41	author	author	NOUN
bracis-33630	305	42	profile	profile	PROPN
bracis-33630	305	43	joão	joão	PROPN
bracis-33630	305	44	gama	gama	PROPN
bracis-33630	305	45	view	view	NOUN
bracis-33630	305	46	author	author	NOUN
bracis-33630	305	47	profile	profile	NOUN
bracis-33630	305	48	search	search	NOUN
bracis-33630	305	49	search	search	NOUN
bracis-33630	305	50	by	by	ADP
bracis-33630	305	51	keyword	keyword	NOUN
bracis-33630	305	52	or	or	CCONJ
bracis-33630	305	53	author	author	NOUN
bracis-33630	305	54	search	search	NOUN
bracis-33630	305	55	navigation	navigation	NOUN
bracis-33630	305	56	find	find	VERB
bracis-33630	305	57	a	a	DET
bracis-33630	305	58	journal	journal	NOUN
bracis-33630	305	59	publish	publish	VERB
bracis-33630	305	60	with	with	ADP
bracis-33630	305	61	us	we	PRON
bracis-33630	305	62	track	track	VERB
bracis-33630	305	63	your	your	PRON
bracis-33630	305	64	research	research	NOUN
bracis-33630	305	65	discover	discover	VERB
bracis-33630	305	66	content	content	NOUN
bracis-33630	305	67	journals	journal	NOUN
bracis-33630	305	68	a	a	DET
bracis-33630	305	69	-	-	PUNCT
bracis-33630	305	70	z	z	NOUN
bracis-33630	305	71	books	book	NOUN
bracis-33630	305	72	a	a	DET
bracis-33630	305	73	-	-	PUNCT
bracis-33630	305	74	z	z	NOUN
bracis-33630	305	75	publish	publish	NOUN
bracis-33630	305	76	with	with	ADP
bracis-33630	305	77	us	us	PROPN
bracis-33630	305	78	journal	journal	PROPN
bracis-33630	305	79	finder	finder	PROPN
bracis-33630	305	80	publish	publish	VERB
bracis-33630	305	81	your	your	PRON
bracis-33630	305	82	research	research	NOUN
bracis-33630	305	83	language	language	NOUN
bracis-33630	305	84	editing	edit	VERB
bracis-33630	305	85	open	open	ADJ
bracis-33630	305	86	access	access	NOUN
bracis-33630	305	87	publishing	publishing	NOUN
bracis-33630	305	88	products	product	NOUN
bracis-33630	305	89	and	and	CCONJ
bracis-33630	305	90	services	service	NOUN
bracis-33630	305	91	our	our	PRON
bracis-33630	305	92	products	product	NOUN
bracis-33630	305	93	librarians	librarian	VERB
bracis-33630	305	94	societies	society	NOUN
bracis-33630	305	95	partners	partner	NOUN
bracis-33630	305	96	and	and	CCONJ
bracis-33630	305	97	advertisers	advertiser	NOUN
bracis-33630	305	98	our	our	PRON
bracis-33630	305	99	brands	brand	NOUN
bracis-33630	305	100	springer	springer	NOUN
bracis-33630	305	101	nature	nature	PROPN
bracis-33630	305	102	portfolio	portfolio	PROPN
bracis-33630	305	103	bmc	bmc	PROPN
bracis-33630	305	104	palgrave	palgrave	PROPN
bracis-33630	305	105	macmillan	macmillan	PROPN
bracis-33630	305	106	apress	apress	PROPN
bracis-33630	305	107	discover	discover	VERB
bracis-33630	305	108	your	your	PRON
bracis-33630	305	109	privacy	privacy	NOUN
bracis-33630	305	110	choices	choice	NOUN
bracis-33630	305	111	/	/	SYM
bracis-33630	305	112	manage	manage	NOUN
bracis-33630	305	113	cookies	cookie	NOUN
bracis-33630	305	114	your	your	PRON
bracis-33630	305	115	us	us	PROPN
bracis-33630	306	1	state	state	NOUN
bracis-33630	306	2	privacy	privacy	NOUN
bracis-33630	306	3	rights	right	NOUN
bracis-33630	306	4	accessibility	accessibility	NOUN
bracis-33630	306	5	statement	statement	NOUN
bracis-33630	306	6	terms	term	NOUN
bracis-33630	306	7	and	and	CCONJ
bracis-33630	306	8	conditions	condition	NOUN
bracis-33630	306	9	privacy	privacy	NOUN
bracis-33630	306	10	policy	policy	NOUN
bracis-33630	306	11	help	help	NOUN
bracis-33630	306	12	and	and	CCONJ
bracis-33630	306	13	support	support	VERB
bracis-33630	306	14	legal	legal	ADJ
bracis-33630	306	15	notice	notice	NOUN
bracis-33630	306	16	cancel	cancel	VERB
bracis-33630	306	17	contracts	contract	NOUN
bracis-33630	306	18	here	here	ADV
bracis-33630	306	19	129.74.145.123	129.74.145.123	NUM
bracis-33630	306	20	hesburgh	hesburgh	PROPN
bracis-33630	306	21	library	library	PROPN
bracis-33630	306	22	er	er	INTJ
bracis-33630	306	23	unit	unit	NOUN
bracis-33630	306	24	(	(	PUNCT
bracis-33630	306	25	3005732405	3005732405	NUM
bracis-33630	306	26	)	)	PUNCT
bracis-33630	306	27	northeast	northeast	ADJ
bracis-33630	306	28	research	research	NOUN
bracis-33630	306	29	libraries	library	NOUN
bracis-33630	306	30	(	(	PUNCT
bracis-33630	306	31	nerl	nerl	PROPN
bracis-33630	306	32	)	)	PUNCT
bracis-33630	306	33	(	(	PUNCT
bracis-33630	306	34	8200828607	8200828607	NUM
bracis-33630	306	35	)	)	PUNCT
bracis-33630	306	36	nerl	nerl	VERB
bracis-33630	306	37	ta	ta	X
bracis-33630	306	38	account	account	NOUN
bracis-33630	306	39	(	(	PUNCT
bracis-33630	306	40	3006206169	3006206169	NUM
bracis-33630	306	41	)	)	PUNCT
bracis-33630	306	42	university	university	NOUN
bracis-33630	306	43	of	of	ADP
bracis-33630	306	44	notre	notre	PROPN
bracis-33630	306	45	dame	dame	PROPN
bracis-33630	306	46	hesburgh	hesburgh	PROPN
bracis-33630	306	47	library	library	NOUN
bracis-33630	306	48	(	(	PUNCT
bracis-33630	306	49	3000184373	3000184373	NUM
bracis-33630	306	50	)	)	PUNCT
bracis-33630	307	1	©	©	ADP
bracis-33630	307	2	2025	2025	NUM
bracis-33630	307	3	springer	springer	NOUN
bracis-33630	307	4	nature	nature	NOUN
