id	sid	tid	token	lemma	pos
2163	1	1	Reference	reference	NN
2163	1	2	Information	Information	NNP
2163	1	3	Extraction	Extraction	NNP
2163	1	4	and	and	CC
2163	1	5	Processing	Processing	NNP
2163	1	6	Using	Using	NNP
2163	1	7	Conditional	Conditional	NNP
2163	1	8	Random	Random	NNP
2163	1	9	Fields	Fields	NNPS
2163	1	10	Tudor	Tudor	NNP
2163	1	11	Groza	Groza	NNP
2163	1	12	,	,	,
2163	1	13	Gunnar	Gunnar	NNP
2163	1	14	AAstrand	AAstrand	NNP
2163	1	15	Grimnes	Grimnes	NNPS
2163	1	16	,	,	,
2163	1	17	and	and	CC
2163	1	18	Siegfried	Siegfried	NNP
2163	1	19	Handschuh	Handschuh	NNP
2163	1	20	REFERENCE	REFERENCE	NNP
2163	1	21	INFORMATION	INFORMATION	VBD
2163	1	22	EXTRACTION	extraction	NN
2163	1	23	AND	and	CC
2163	1	24	PROCESSING	processing	NN
2163	1	25	|GROZA	|groza	NN
2163	1	26	,	,	,
2163	1	27	GRIMNES	GRIMNES	NNP
2163	1	28	,	,	,
2163	1	29	AND	and	CC
2163	1	30	HANDSCHUH	HANDSCHUH	NNP
2163	1	31	6	6	CD
2163	1	32	ABSTRACT	ABSTRACT	NNP
2163	1	33	Fostering	foster	VBG
2163	1	34	both	both	CC
2163	1	35	the	the	DT
2163	1	36	creation	creation	NN
2163	1	37	and	and	CC
2163	1	38	the	the	DT
2163	1	39	linking	linking	NN
2163	1	40	of	of	IN
2163	1	41	data	datum	NNS
2163	1	42	with	with	IN
2163	1	43	the	the	DT
2163	1	44	scope	scope	NN
2163	1	45	of	of	IN
2163	1	46	supporting	support	VBG
2163	1	47	the	the	DT
2163	1	48	growth	growth	NN
2163	1	49	of	of	IN
2163	1	50	the	the	DT
2163	1	51	Linked	Linked	NNP
2163	1	52	Data	Data	NNP
2163	1	53	Web	Web	NNP
2163	1	54	requires	require	VBZ
2163	1	55	us	-PRON-	PRP
2163	1	56	to	to	TO
2163	1	57	improve	improve	VB
2163	1	58	the	the	DT
2163	1	59	acquisition	acquisition	NN
2163	1	60	and	and	CC
2163	1	61	extraction	extraction	NN
2163	1	62	mechanisms	mechanism	NNS
2163	1	63	of	of	IN
2163	1	64	the	the	DT
2163	1	65	underlying	underlie	VBG
2163	1	66	semantic	semantic	JJ
2163	1	67	metadata	metadata	NN
2163	1	68	.	.	.
2163	2	1	This	this	DT
2163	2	2	is	be	VBZ
2163	2	3	particularly	particularly	RB
2163	2	4	important	important	JJ
2163	2	5	for	for	IN
2163	2	6	the	the	DT
2163	2	7	scientific	scientific	JJ
2163	2	8	publishing	publishing	NN
2163	2	9	domain	domain	NN
2163	2	10	,	,	,
2163	2	11	where	where	WRB
2163	2	12	currently	currently	RB
2163	2	13	most	most	JJS
2163	2	14	of	of	IN
2163	2	15	the	the	DT
2163	2	16	datasets	dataset	NNS
2163	2	17	are	be	VBP
2163	2	18	being	be	VBG
2163	2	19	created	create	VBN
2163	2	20	in	in	IN
2163	2	21	an	an	DT
2163	2	22	author	author	NN
2163	2	23	-	-	HYPH
2163	2	24	driven	drive	VBN
2163	2	25	,	,	,
2163	2	26	manual	manual	JJ
2163	2	27	manner	manner	NN
2163	2	28	.	.	.
2163	3	1	In	in	IN
2163	3	2	addition	addition	NN
2163	3	3	,	,	,
2163	3	4	such	such	JJ
2163	3	5	datasets	dataset	NNS
2163	3	6	capture	capture	VBP
2163	3	7	only	only	RB
2163	3	8	fragments	fragment	NNS
2163	3	9	of	of	IN
2163	3	10	the	the	DT
2163	3	11	complete	complete	JJ
2163	3	12	metadata	metadata	NN
2163	3	13	,	,	,
2163	3	14	omitting	omit	VBG
2163	3	15	usually	usually	RB
2163	3	16	,	,	,
2163	3	17	important	important	JJ
2163	3	18	elements	element	NNS
2163	3	19	such	such	JJ
2163	3	20	as	as	IN
2163	3	21	the	the	DT
2163	3	22	references	reference	NNS
2163	3	23	,	,	,
2163	3	24	although	although	IN
2163	3	25	they	-PRON-	PRP
2163	3	26	represent	represent	VBP
2163	3	27	valuable	valuable	JJ
2163	3	28	information	information	NN
2163	3	29	.	.	.
2163	4	1	In	in	IN
2163	4	2	this	this	DT
2163	4	3	paper	paper	NN
2163	4	4	we	-PRON-	PRP
2163	4	5	present	present	VBP
2163	4	6	an	an	DT
2163	4	7	approach	approach	NN
2163	4	8	that	that	WDT
2163	4	9	aims	aim	VBZ
2163	4	10	at	at	IN
2163	4	11	dealing	deal	VBG
2163	4	12	with	with	IN
2163	4	13	this	this	DT
2163	4	14	aspect	aspect	NN
2163	4	15	of	of	IN
2163	4	16	extraction	extraction	NN
2163	4	17	and	and	CC
2163	4	18	processing	processing	NN
2163	4	19	of	of	IN
2163	4	20	reference	reference	NN
2163	4	21	information	information	NN
2163	4	22	.	.	.
2163	5	1	The	the	DT
2163	5	2	experimental	experimental	JJ
2163	5	3	evaluation	evaluation	NN
2163	5	4	shows	show	VBZ
2163	5	5	that	that	IN
2163	5	6	,	,	,
2163	5	7	currently	currently	RB
2163	5	8	,	,	,
2163	5	9	our	-PRON-	PRP$
2163	5	10	solution	solution	NN
2163	5	11	handles	handle	VBZ
2163	5	12	very	very	RB
2163	5	13	well	well	RB
2163	5	14	diverse	diverse	JJ
2163	5	15	types	type	NNS
2163	5	16	of	of	IN
2163	5	17	reference	reference	NN
2163	5	18	format	format	NN
2163	5	19	,	,	,
2163	5	20	thus	thus	RB
2163	5	21	making	make	VBG
2163	5	22	it	-PRON-	PRP
2163	5	23	usable	usable	JJ
2163	5	24	for	for	IN
2163	5	25	,	,	,
2163	5	26	or	or	CC
2163	5	27	adaptable	adaptable	JJ
2163	5	28	to	to	IN
2163	5	29	,	,	,
2163	5	30	any	any	DT
2163	5	31	area	area	NN
2163	5	32	of	of	IN
2163	5	33	scientific	scientific	JJ
2163	5	34	publishing	publishing	NN
2163	5	35	.	.	.
2163	6	1	1	1	LS
2163	6	2	.	.	.
2163	7	1	INTRODUCTION	introduction	VB
2163	7	2	The	the	DT
2163	7	3	progressive	progressive	JJ
2163	7	4	adoption	adoption	NN
2163	7	5	of	of	IN
2163	7	6	Semantic	Semantic	NNP
2163	7	7	Web	web	NN
2163	7	8	1	1	CD
2163	7	9	techniques	technique	NNS
2163	7	10	resulted	result	VBN
2163	7	11	in	in	IN
2163	7	12	the	the	DT
2163	7	13	creation	creation	NN
2163	7	14	of	of	IN
2163	7	15	a	a	DT
2163	7	16	series	series	NN
2163	7	17	of	of	IN
2163	7	18	datasets	dataset	NNS
2163	7	19	connected	connect	VBN
2163	7	20	by	by	IN
2163	7	21	the	the	DT
2163	7	22	Linked	Linked	NNP
2163	7	23	Data	Data	NNP
2163	7	24	2	2	CD
2163	7	25	initiative	initiative	NN
2163	7	26	,	,	,
2163	7	27	and	and	CC
2163	7	28	via	via	IN
2163	7	29	the	the	DT
2163	7	30	Linked	Linked	NNP
2163	7	31	Data	Data	NNP
2163	7	32	principles	principle	NNS
2163	7	33	,	,	,
2163	7	34	into	into	IN
2163	7	35	a	a	DT
2163	7	36	universal	universal	JJ
2163	7	37	Web	web	NN
2163	7	38	of	of	IN
2163	7	39	Linked	Linked	NNP
2163	7	40	Data	Data	NNP
2163	7	41	.	.	.
2163	8	1	In	in	IN
2163	8	2	order	order	NN
2163	8	3	to	to	TO
2163	8	4	foster	foster	VB
2163	8	5	the	the	DT
2163	8	6	continuous	continuous	JJ
2163	8	7	growth	growth	NN
2163	8	8	of	of	IN
2163	8	9	this	this	DT
2163	8	10	Linked	Linked	NNP
2163	8	11	Data	Data	NNP
2163	8	12	Web	Web	NNP
2163	8	13	,	,	,
2163	8	14	we	-PRON-	PRP
2163	8	15	need	need	VBP
2163	8	16	to	to	TO
2163	8	17	improve	improve	VB
2163	8	18	the	the	DT
2163	8	19	acquisition	acquisition	NN
2163	8	20	and	and	CC
2163	8	21	extraction	extraction	NN
2163	8	22	mechanisms	mechanism	NNS
2163	8	23	of	of	IN
2163	8	24	the	the	DT
2163	8	25	underlying	underlie	VBG
2163	8	26	semantic	semantic	JJ
2163	8	27	metadata	metadata	NN
2163	8	28	.	.	.
2163	9	1	Unfortunately	unfortunately	RB
2163	9	2	,	,	,
2163	9	3	the	the	DT
2163	9	4	scientific	scientific	JJ
2163	9	5	publishing	publishing	NN
2163	9	6	domain	domain	NN
2163	9	7	,	,	,
2163	9	8	a	a	DT
2163	9	9	domain	domain	NN
2163	9	10	with	with	IN
2163	9	11	an	an	DT
2163	9	12	enormous	enormous	JJ
2163	9	13	potential	potential	NN
2163	9	14	for	for	IN
2163	9	15	generating	generate	VBG
2163	9	16	large	large	JJ
2163	9	17	amounts	amount	NNS
2163	9	18	of	of	IN
2163	9	19	Linked	Linked	NNP
2163	9	20	Data	Data	NNP
2163	9	21	,	,	,
2163	9	22	still	still	RB
2163	9	23	promotes	promote	VBZ
2163	9	24	trivial	trivial	JJ
2163	9	25	mechanisms	mechanism	NNS
2163	9	26	for	for	IN
2163	9	27	producing	produce	VBG
2163	9	28	semantic	semantic	JJ
2163	9	29	metadata	metadata	NN
2163	9	30	.	.	.
2163	10	1	3	3	LS
2163	10	2	As	as	IN
2163	10	3	an	an	DT
2163	10	4	illustration	illustration	NN
2163	10	5	,	,	,
2163	10	6	the	the	DT
2163	10	7	metadata	metadata	NN
2163	10	8	acquisition	acquisition	NN
2163	10	9	process	process	NN
2163	10	10	of	of	IN
2163	10	11	the	the	DT
2163	10	12	Semantic	Semantic	NNP
2163	10	13	Web	Web	NNP
2163	10	14	Dog	Dog	NNP
2163	10	15	Food	Food	NNP
2163	10	16	Server	Server	NNP
2163	10	17	,	,	,
2163	10	18	4	4	CD
2163	10	19	the	the	DT
2163	10	20	main	main	JJ
2163	10	21	Linked	Linked	NNP
2163	10	22	Data	Data	NNP
2163	10	23	publication	publication	NN
2163	10	24	repository	repository	NN
2163	10	25	available	available	JJ
2163	10	26	on	on	IN
2163	10	27	the	the	DT
2163	10	28	Web	web	NN
2163	10	29	,	,	,
2163	10	30	consists	consist	VBZ
2163	10	31	of	of	IN
2163	10	32	two	two	CD
2163	10	33	steps	step	NNS
2163	10	34	:	:	:
2163	10	35			.
2163	10	36	the	the	DT
2163	10	37	authors	author	NNS
2163	10	38	manually	manually	RB
2163	10	39	fill	fill	VBP
2163	10	40	-	-	HYPH
2163	10	41	in	in	RP
2163	10	42	submission	submission	NN
2163	10	43	forms	form	NNS
2163	10	44	corresponding	correspond	VBG
2163	10	45	to	to	IN
2163	10	46	different	different	JJ
2163	10	47	publishing	publishing	NN
2163	10	48	venues	venue	NNS
2163	10	49	(	(	-LRB-
2163	10	50	e.g.	e.g.	RB
2163	10	51	,	,	,
2163	10	52	conferences	conference	NNS
2163	10	53	or	or	CC
2163	10	54	workshops	workshop	NNS
2163	10	55	)	)	-RRB-
2163	10	56	,	,	,
2163	10	57	with	with	IN
2163	10	58	the	the	DT
2163	10	59	resulting	resulting	NN
2163	10	60	(	(	-LRB-
2163	10	61	usually	usually	RB
2163	10	62	XML	xml	NN
2163	10	63	)	)	-RRB-
2163	10	64	information	information	NN
2163	10	65	being	be	VBG
2163	10	66	transformed	transform	VBN
2163	10	67	via	via	IN
2163	10	68	scripts	script	NNS
2163	10	69	into	into	IN
2163	10	70	semantic	semantic	JJ
2163	10	71	metadata	metadata	NN
2163	10	72	,	,	,
2163	10	73	and	and	CC
2163	10	74			NFP
2163	10	75	the	the	DT
2163	10	76	entity	entity	NN
2163	10	77	URIs	uri	NNS
2163	10	78	(	(	-LRB-
2163	10	79	i.e.	i.e.	FW
2163	10	80	,	,	,
2163	10	81	authors	author	NNS
2163	10	82	and	and	CC
2163	10	83	publications	publication	NNS
2163	10	84	)	)	-RRB-
2163	10	85	present	present	JJ
2163	10	86	in	in	IN
2163	10	87	this	this	DT
2163	10	88	semantic	semantic	JJ
2163	10	89	metadata	metadata	NN
2163	10	90	are	be	VBP
2163	10	91	then	then	RB
2163	10	92	manually	manually	RB
2163	10	93	mapped	map	VBN
2163	10	94	to	to	IN
2163	10	95	existing	exist	VBG
2163	10	96	Web	web	NN
2163	10	97	URIs	uri	NNS
2163	10	98	for	for	IN
2163	10	99	linking	linking	NN
2163	10	100	/	/	SYM
2163	10	101	consolidation	consolidation	NN
2163	10	102	purposes	purpose	NNS
2163	10	103	.	.	.
2163	11	1	Tudor	Tudor	NNP
2163	11	2	Groza	Groza	NNP
2163	11	3	(	(	-LRB-
2163	11	4	tudor.groza@uq.edu.au	tudor.groza@uq.edu.au	NN
2163	11	5	)	)	-RRB-
2163	11	6	is	be	VBZ
2163	11	7	Postdoctoral	Postdoctoral	NNP
2163	11	8	Research	Research	NNP
2163	11	9	Fellow	Fellow	NNP
2163	11	10	,	,	,
2163	11	11	School	School	NNP
2163	11	12	of	of	IN
2163	11	13	Information	Information	NNP
2163	11	14	Technology	Technology	NNP
2163	11	15	and	and	CC
2163	11	16	Electrical	Electrical	NNP
2163	11	17	Engineering	Engineering	NNP
2163	11	18	,	,	,
2163	11	19	University	University	NNP
2163	11	20	of	of	IN
2163	11	21	Queensland	Queensland	NNP
2163	11	22	,	,	,
2163	11	23	Gunnar	Gunnar	NNP
2163	11	24	AAstrand	AAstrand	NNP
2163	11	25	Grimnes	Grimnes	NNPS
2163	11	26	(	(	-LRB-
2163	11	27	grimnes@dfki.uni-kl.de	grimnes@dfki.uni-kl.de	NNP
2163	11	28	)	)	-RRB-
2163	11	29	is	be	VBZ
2163	11	30	Researcher	Researcher	NNP
2163	11	31	,	,	,
2163	11	32	German	German	NNP
2163	11	33	Research	Research	NNP
2163	11	34	Center	Center	NNP
2163	11	35	for	for	IN
2163	11	36	Artificial	Artificial	NNP
2163	11	37	Intelligence	Intelligence	NNP
2163	11	38	(	(	-LRB-
2163	11	39	DFKI	DFKI	NNP
2163	11	40	)	)	-RRB-
2163	11	41	GmbH	GmbH	NNP
2163	11	42	,	,	,
2163	11	43	Kaiserslautern	Kaiserslautern	NNP
2163	11	44	,	,	,
2163	11	45	Germany	Germany	NNP
2163	11	46	,	,	,
2163	11	47	Siegfried	Siegfried	NNP
2163	11	48	Handschuh	Handschuh	NNP
2163	11	49	(	(	-LRB-
2163	11	50	msiegfried.handschuh@deri.org	msiegfried.handschuh@deri.org	NNP
2163	11	51	)	)	-RRB-
2163	11	52	is	be	VBZ
2163	11	53	Senior	Senior	NNP
2163	11	54	Lecturer	Lecturer	NNP
2163	11	55	/	/	SYM
2163	11	56	Associate	Associate	NNP
2163	11	57	Professor	Professor	NNP
2163	11	58	,	,	,
2163	11	59	National	National	NNP
2163	11	60	University	University	NNP
2163	11	61	of	of	IN
2163	11	62	Ireland	Ireland	NNP
2163	11	63	,	,	,
2163	11	64	Galway	Galway	NNP
2163	11	65	,	,	,
2163	11	66	Ireland	Ireland	NNP
2163	11	67	.	.	.
2163	12	1	mailto:tudor.groza@uq.edu.au	mailto:tudor.groza@uq.edu.au	NNP
2163	12	2	mailto:grimnes@dfki.uni-kl.de	mailto:grimnes@dfki.uni-kl.de	NNP
2163	12	3	mailto:msiegfried.handschuh@deri.org	mailto:msiegfried.handschuh@deri.org	NNP
2163	12	4	INFORMATION	INFORMATION	VBD
2163	12	5	TECHNOLOGY	technology	NN
2163	12	6	AND	and	CC
2163	12	7	LIBRARIES	LIBRARIES	NNP
2163	12	8	|	|	NNP
2163	12	9	JUNE	JUNE	NNP
2163	12	10	2012	2012	CD
2163	12	11	7	7	CD
2163	12	12	Moreover	moreover	RB
2163	12	13	,	,	,
2163	12	14	independent	independent	JJ
2163	12	15	of	of	IN
2163	12	16	the	the	DT
2163	12	17	creation	creation	NN
2163	12	18	/	/	SYM
2163	12	19	acquisition	acquisition	NN
2163	12	20	process	process	NN
2163	12	21	,	,	,
2163	12	22	one	one	CD
2163	12	23	particular	particular	JJ
2163	12	24	component	component	NN
2163	12	25	of	of	IN
2163	12	26	the	the	DT
2163	12	27	publication	publication	NN
2163	12	28	metadata	metadata	NN
2163	12	29	,	,	,
2163	12	30	i.e.	i.e.	FW
2163	12	31	,	,	,
2163	12	32	the	the	DT
2163	12	33	reference	reference	NN
2163	12	34	information	information	NN
2163	12	35	,	,	,
2163	12	36	is	be	VBZ
2163	12	37	almost	almost	RB
2163	12	38	constantly	constantly	RB
2163	12	39	neglected	neglect	VBN
2163	12	40	.	.	.
2163	13	1	The	the	DT
2163	13	2	reason	reason	NN
2163	13	3	is	be	VBZ
2163	13	4	mainly	mainly	RB
2163	13	5	the	the	DT
2163	13	6	amount	amount	NN
2163	13	7	of	of	IN
2163	13	8	work	work	NN
2163	13	9	required	require	VBN
2163	13	10	to	to	TO
2163	13	11	manually	manually	RB
2163	13	12	create	create	VB
2163	13	13	it	-PRON-	PRP
2163	13	14	,	,	,
2163	13	15	or	or	CC
2163	13	16	the	the	DT
2163	13	17	complexity	complexity	NN
2163	13	18	of	of	IN
2163	13	19	the	the	DT
2163	13	20	task	task	NN
2163	13	21	,	,	,
2163	13	22	in	in	IN
2163	13	23	the	the	DT
2163	13	24	case	case	NN
2163	13	25	of	of	IN
2163	13	26	automatic	automatic	JJ
2163	13	27	extraction	extraction	NN
2163	13	28	.	.	.
2163	14	1	As	as	IN
2163	14	2	a	a	DT
2163	14	3	result	result	NN
2163	14	4	,	,	,
2163	14	5	currently	currently	RB
2163	14	6	,	,	,
2163	14	7	there	there	EX
2163	14	8	are	be	VBP
2163	14	9	no	no	DT
2163	14	10	datasets	dataset	NNS
2163	14	11	in	in	IN
2163	14	12	the	the	DT
2163	14	13	Linked	Linked	NNP
2163	14	14	Data	Data	NNP
2163	14	15	Web	web	NN
2163	14	16	exposing	expose	VBG
2163	14	17	reference	reference	NN
2163	14	18	information	information	NN
2163	14	19	,	,	,
2163	14	20	while	while	IN
2163	14	21	the	the	DT
2163	14	22	number	number	NN
2163	14	23	of	of	IN
2163	14	24	digital	digital	JJ
2163	14	25	libraries	library	NNS
2163	14	26	providing	provide	VBG
2163	14	27	search	search	NN
2163	14	28	and	and	CC
2163	14	29	link	link	VBP
2163	14	30	functionality	functionality	NN
2163	14	31	over	over	IN
2163	14	32	references	reference	NNS
2163	14	33	is	be	VBZ
2163	14	34	rather	rather	RB
2163	14	35	limited	limited	JJ
2163	14	36	.	.	.
2163	15	1	This	this	DT
2163	15	2	is	be	VBZ
2163	15	3	quite	quite	PDT
2163	15	4	a	a	DT
2163	15	5	problematic	problematic	JJ
2163	15	6	gap	gap	NN
2163	15	7	if	if	IN
2163	15	8	we	-PRON-	PRP
2163	15	9	consider	consider	VBP
2163	15	10	the	the	DT
2163	15	11	amount	amount	NN
2163	15	12	of	of	IN
2163	15	13	information	information	NN
2163	15	14	provided	provide	VBN
2163	15	15	by	by	IN
2163	15	16	references	reference	NNS
2163	15	17	and	and	CC
2163	15	18	their	-PRON-	PRP$
2163	15	19	foundational	foundational	JJ
2163	15	20	support	support	NN
2163	15	21	for	for	IN
2163	15	22	other	other	JJ
2163	15	23	application	application	NN
2163	15	24	techniques	technique	NNS
2163	15	25	that	that	WDT
2163	15	26	bring	bring	VBP
2163	15	27	value	value	NN
2163	15	28	to	to	IN
2163	15	29	researchers	researcher	NNS
2163	15	30	and	and	CC
2163	15	31	librarians	librarian	NNS
2163	15	32	,	,	,
2163	15	33	such	such	JJ
2163	15	34	as	as	IN
2163	15	35	citation	citation	NN
2163	15	36	analysis	analysis	NN
2163	15	37	and	and	CC
2163	15	38	citation	citation	NN
2163	15	39	metrics	metric	NNS
2163	15	40	,	,	,
2163	15	41	tracking	track	VBG
2163	15	42	temporal	temporal	JJ
2163	15	43	author	author	NN
2163	15	44	-	-	HYPH
2163	15	45	topic	topic	NN
2163	15	46	evolution	evolution	NN
2163	15	47	5	5	CD
2163	15	48	or	or	CC
2163	15	49	co	co	NN
2163	15	50	-	-	JJ
2163	15	51	authorship	authorship	JJ
2163	15	52	graph	graph	NN
2163	15	53	analysis	analysis	NN
2163	15	54	.	.	.
2163	16	1	6,7	6,7	LS
2163	16	2	In	in	IN
2163	16	3	this	this	DT
2163	16	4	paper	paper	NN
2163	16	5	we	-PRON-	PRP
2163	16	6	focus	focus	VBP
2163	16	7	on	on	IN
2163	16	8	the	the	DT
2163	16	9	first	first	JJ
2163	16	10	of	of	IN
2163	16	11	the	the	DT
2163	16	12	above	above	RB
2163	16	13	-	-	HYPH
2163	16	14	mentioned	mention	VBN
2163	16	15	steps	step	NNS
2163	16	16	,	,	,
2163	16	17	i.e.	i.e.	FW
2163	16	18	,	,	,
2163	16	19	providing	provide	VBG
2163	16	20	the	the	DT
2163	16	21	underlying	underlie	VBG
2163	16	22	mechanisms	mechanism	NNS
2163	16	23	for	for	IN
2163	16	24	automatic	automatic	JJ
2163	16	25	extraction	extraction	NN
2163	16	26	of	of	IN
2163	16	27	reference	reference	NN
2163	16	28	metadata	metadata	NN
2163	16	29	.	.	.
2163	17	1	We	-PRON-	PRP
2163	17	2	devise	devise	VBP
2163	17	3	a	a	DT
2163	17	4	solution	solution	NN
2163	17	5	that	that	WDT
2163	17	6	enables	enable	VBZ
2163	17	7	extraction	extraction	NN
2163	17	8	and	and	CC
2163	17	9	chunking	chunking	NN
2163	17	10	of	of	IN
2163	17	11	references	reference	NNS
2163	17	12	using	use	VBG
2163	17	13	Conditional	Conditional	NNP
2163	17	14	Random	Random	NNP
2163	17	15	Fields	Fields	NNPS
2163	17	16	(	(	-LRB-
2163	17	17	CRF	CRF	NNP
2163	17	18	)	)	-RRB-
2163	17	19	.	.	.
2163	18	1	8	8	CD
2163	18	2	The	the	DT
2163	18	3	resulting	result	VBG
2163	18	4	metadata	metadata	NN
2163	18	5	can	can	MD
2163	18	6	then	then	RB
2163	18	7	be	be	VB
2163	18	8	easily	easily	RB
2163	18	9	transformed	transform	VBN
2163	18	10	into	into	IN
2163	18	11	semantic	semantic	JJ
2163	18	12	metadata	metadata	NN
2163	18	13	adhering	adhere	VBG
2163	18	14	to	to	IN
2163	18	15	particular	particular	JJ
2163	18	16	schemas	schema	NNS
2163	18	17	via	via	IN
2163	18	18	scripts	script	NNS
2163	18	19	,	,	,
2163	18	20	the	the	DT
2163	18	21	added	add	VBN
2163	18	22	value	value	NN
2163	18	23	being	be	VBG
2163	18	24	the	the	DT
2163	18	25	exclusion	exclusion	NN
2163	18	26	of	of	IN
2163	18	27	the	the	DT
2163	18	28	manual	manual	JJ
2163	18	29	author	author	NN
2163	18	30	-	-	HYPH
2163	18	31	driven	drive	VBN
2163	18	32	creation	creation	NN
2163	18	33	step	step	NN
2163	18	34	from	from	IN
2163	18	35	the	the	DT
2163	18	36	process	process	NN
2163	18	37	.	.	.
2163	19	1	From	from	IN
2163	19	2	the	the	DT
2163	19	3	domain	domain	NN
2163	19	4	perspective	perspective	NN
2163	19	5	,	,	,
2163	19	6	we	-PRON-	PRP
2163	19	7	focus	focus	VBP
2163	19	8	on	on	IN
2163	19	9	computer	computer	NN
2163	19	10	science	science	NN
2163	19	11	and	and	CC
2163	19	12	health	health	NN
2163	19	13	sciences	science	NNS
2163	19	14	only	only	RB
2163	19	15	because	because	IN
2163	19	16	these	these	DT
2163	19	17	domains	domain	NNS
2163	19	18	have	have	VBP
2163	19	19	representative	representative	JJ
2163	19	20	datasets	dataset	NNS
2163	19	21	that	that	WDT
2163	19	22	can	can	MD
2163	19	23	be	be	VB
2163	19	24	used	use	VBN
2163	19	25	for	for	IN
2163	19	26	evaluation	evaluation	NN
2163	19	27	and	and	CC
2163	19	28	hence	hence	RB
2163	19	29	enable	enable	VB
2163	19	30	comparison	comparison	NN
2163	19	31	against	against	IN
2163	19	32	similar	similar	JJ
2163	19	33	approaches	approach	NNS
2163	19	34	.	.	.
2163	20	1	However	however	RB
2163	20	2	,	,	,
2163	20	3	we	-PRON-	PRP
2163	20	4	believe	believe	VBP
2163	20	5	that	that	IN
2163	20	6	our	-PRON-	PRP$
2163	20	7	model	model	NN
2163	20	8	can	can	MD
2163	20	9	be	be	VB
2163	20	10	applied	apply	VBN
2163	20	11	also	also	RB
2163	20	12	in	in	IN
2163	20	13	domains	domain	NNS
2163	20	14	such	such	JJ
2163	20	15	as	as	IN
2163	20	16	digital	digital	JJ
2163	20	17	humanities	humanity	NNS
2163	20	18	or	or	CC
2163	20	19	social	social	JJ
2163	20	20	sciences	science	NNS
2163	20	21	,	,	,
2163	20	22	and	and	CC
2163	20	23	we	-PRON-	PRP
2163	20	24	intend	intend	VBP
2163	20	25	,	,	,
2163	20	26	in	in	IN
2163	20	27	the	the	DT
2163	20	28	near	near	JJ
2163	20	29	future	future	NN
2163	20	30	,	,	,
2163	20	31	to	to	TO
2163	20	32	build	build	VB
2163	20	33	a	a	DT
2163	20	34	corresponding	corresponding	JJ
2163	20	35	corpus	corpus	NN
2163	20	36	that	that	WDT
2163	20	37	would	would	MD
2163	20	38	allow	allow	VB
2163	20	39	us	-PRON-	PRP
2163	20	40	to	to	TO
2163	20	41	test	test	VB
2163	20	42	and	and	CC
2163	20	43	adapt	adapt	VB
2163	20	44	(	(	-LRB-
2163	20	45	if	if	IN
2163	20	46	necessary	necessary	JJ
2163	20	47	)	)	-RRB-
2163	20	48	our	-PRON-	PRP$
2163	20	49	solution	solution	NN
2163	20	50	to	to	IN
2163	20	51	these	these	DT
2163	20	52	domains	domain	NNS
2163	20	53	.	.	.
2163	21	1	Figure	figure	NN
2163	21	2	1	1	CD
2163	21	3	.	.	.
2163	22	1	Examples	example	NNS
2163	22	2	of	of	IN
2163	22	3	Chunked	Chunked	NNP
2163	22	4	and	and	CC
2163	22	5	Labeled	Labeled	NNP
2163	22	6	Reference	Reference	NNP
2163	22	7	Strings	Strings	NNP
2163	22	8	Reference	Reference	NNP
2163	22	9	chunking	chunking	NN
2163	22	10	represents	represent	VBZ
2163	22	11	the	the	DT
2163	22	12	process	process	NN
2163	22	13	of	of	IN
2163	22	14	label	label	NNP
2163	22	15	sequencing	sequence	VBG
2163	22	16	a	a	DT
2163	22	17	reference	reference	NN
2163	22	18	string	string	NN
2163	22	19	,	,	,
2163	22	20	i.e.	i.e.	FW
2163	22	21	,	,	,
2163	22	22	tagging	tag	VBG
2163	22	23	the	the	DT
2163	22	24	parts	part	NNS
2163	22	25	of	of	IN
2163	22	26	the	the	DT
2163	22	27	reference	reference	NN
2163	22	28	containing	contain	VBG
2163	22	29	the	the	DT
2163	22	30	authors	author	NNS
2163	22	31	,	,	,
2163	22	32	the	the	DT
2163	22	33	title	title	NN
2163	22	34	,	,	,
2163	22	35	the	the	DT
2163	22	36	publication	publication	NN
2163	22	37	venue	venue	NN
2163	22	38	,	,	,
2163	22	39	etc	etc	FW
2163	22	40	.	.	.
2163	23	1	The	the	DT
2163	23	2	main	main	JJ
2163	23	3	issue	issue	NN
2163	23	4	associated	associate	VBN
2163	23	5	with	with	IN
2163	23	6	this	this	DT
2163	23	7	task	task	NN
2163	23	8	is	be	VBZ
2163	23	9	the	the	DT
2163	23	10	lack	lack	NN
2163	23	11	of	of	IN
2163	23	12	uniformity	uniformity	NN
2163	23	13	in	in	IN
2163	23	14	the	the	DT
2163	23	15	reference	reference	NN
2163	23	16	representation	representation	NN
2163	23	17	.	.	.
2163	24	1	Figure	figure	NN
2163	24	2	1	1	CD
2163	24	3	presents	present	VBZ
2163	24	4	three	three	CD
2163	24	5	examples	example	NNS
2163	24	6	of	of	IN
2163	24	7	chunked	chunked	JJ
2163	24	8	and	and	CC
2163	24	9	labeled	label	VBN
2163	24	10	reference	reference	NN
2163	24	11	strings	string	NNS
2163	24	12	.	.	.
2163	25	1	One	one	PRP
2163	25	2	can	can	MD
2163	25	3	not	not	RB
2163	25	4	infer	infer	VB
2163	25	5	generic	generic	JJ
2163	25	6	patterns	pattern	NNS
2163	25	7	for	for	IN
2163	25	8	all	all	DT
2163	25	9	types	type	NNS
2163	25	10	of	of	IN
2163	25	11	references	reference	NNS
2163	25	12	.	.	.
2163	26	1	For	for	IN
2163	26	2	example	example	NN
2163	26	3	,	,	,
2163	26	4	the	the	DT
2163	26	5	year	year	NN
2163	26	6	(	(	-LRB-
2163	26	7	or	or	CC
2163	26	8	date	date	NN
2163	26	9	)	)	-RRB-
2163	26	10	of	of	IN
2163	26	11	some	some	DT
2163	26	12	of	of	IN
2163	26	13	the	the	DT
2163	26	14	references	reference	NNS
2163	26	15	of	of	IN
2163	26	16	this	this	DT
2163	26	17	paper	paper	NN
2163	26	18	are	be	VBP
2163	26	19	similar	similar	JJ
2163	26	20	to	to	IN
2163	26	21	example	example	VB
2163	26	22	2	2	CD
2163	26	23	from	from	IN
2163	26	24	the	the	DT
2163	26	25	figure	figure	NN
2163	26	26	,	,	,
2163	26	27	i.e.	i.e.	FW
2163	26	28	,	,	,
2163	26	29	they	-PRON-	PRP
2163	26	30	are	be	VBP
2163	26	31	located	locate	VBN
2163	26	32	at	at	IN
2163	26	33	the	the	DT
2163	26	34	very	very	JJ
2163	26	35	end	end	NN
2163	26	36	of	of	IN
2163	26	37	the	the	DT
2163	26	38	reference	reference	NN
2163	26	39	string	string	NN
2163	26	40	.	.	.
2163	27	1	Unfortunately	unfortunately	RB
2163	27	2	,	,	,
2163	27	3	this	this	DT
2163	27	4	does	do	VBZ
2163	27	5	not	not	RB
2163	27	6	hold	hold	VB
2163	27	7	for	for	IN
2163	27	8	some	some	DT
2163	27	9	journal	journal	NN
2163	27	10	reference	reference	NN
2163	27	11	formats	format	NNS
2163	27	12	,	,	,
2163	27	13	such	such	JJ
2163	27	14	as	as	IN
2163	27	15	the	the	DT
2163	27	16	one	one	CD
2163	27	17	presented	present	VBN
2163	27	18	in	in	IN
2163	27	19	example	example	NN
2163	27	20	1	1	CD
2163	27	21	.	.	.
2163	28	1	And	and	CC
2163	28	2	at	at	IN
2163	28	3	the	the	DT
2163	28	4	same	same	JJ
2163	28	5	time	time	NN
2163	28	6	,	,	,
2163	28	7	the	the	DT
2163	28	8	actual	actual	JJ
2163	28	9	date	date	NN
2163	28	10	might	may	MD
2163	28	11	not	not	RB
2163	28	12	comprise	comprise	VB
2163	28	13	only	only	RB
2163	28	14	the	the	DT
2163	28	15	year	year	NN
2163	28	16	,	,	,
2163	28	17	but	but	CC
2163	28	18	also	also	RB
2163	28	19	the	the	DT
2163	28	20	month	month	NN
2163	28	21	(	(	-LRB-
2163	28	22	and	and	CC
2163	28	23	even	even	RB
2163	28	24	day	day	NN
2163	28	25	)	)	-RRB-
2163	28	26	.	.	.
2163	29	1	In	in	IN
2163	29	2	addition	addition	NN
2163	29	3	to	to	IN
2163	29	4	the	the	DT
2163	29	5	placement	placement	NN
2163	29	6	of	of	IN
2163	29	7	the	the	DT
2163	29	8	particular	particular	JJ
2163	29	9	types	type	NNS
2163	29	10	of	of	IN
2163	29	11	tokens	token	NNS
2163	29	12	within	within	IN
2163	29	13	the	the	DT
2163	29	14	reference	reference	NN
2163	29	15	string	string	NN
2163	29	16	,	,	,
2163	29	17	one	one	CD
2163	29	18	of	of	IN
2163	29	19	the	the	DT
2163	29	20	major	major	JJ
2163	29	21	concerns	concern	NNS
2163	29	22	when	when	WRB
2163	29	23	labeling	label	VBG
2163	29	24	these	these	DT
2163	29	25	types	type	NNS
2163	29	26	of	of	IN
2163	29	27	tokens	token	NNS
2163	29	28	is	be	VBZ
2163	29	29	disambiguation	disambiguation	NN
2163	29	30	.	.	.
2163	30	1	Generally	generally	RB
2163	30	2	,	,	,
2163	30	3	there	there	EX
2163	30	4	are	be	VBP
2163	30	5	three	three	CD
2163	30	6	categories	category	NNS
2163	30	7	of	of	IN
2163	30	8	ambiguous	ambiguous	JJ
2163	30	9	elements	element	NNS
2163	30	10	:	:	:
2163	30	11	REFERENCE	reference	NN
2163	30	12	INFORMATION	INFORMATION	NNS
2163	30	13	EXTRACTION	extraction	NN
2163	30	14	AND	and	CC
2163	30	15	PROCESSING	processing	NN
2163	30	16	|GROZA	|groza	NN
2163	30	17	,	,	,
2163	30	18	GRIMNES	GRIMNES	NNP
2163	30	19	,	,	,
2163	30	20	AND	and	CC
2163	30	21	HANDSCHUH	handschuh	NN
2163	30	22	8	8	CD
2163	30	23			NN
2163	30	24	names	name	NNS
2163	30	25	—	—	:
2163	30	26	can	can	MD
2163	30	27	act	act	VB
2163	30	28	as	as	IN
2163	30	29	authors	author	NNS
2163	30	30	,	,	,
2163	30	31	editors	editor	NNS
2163	30	32	,	,	,
2163	30	33	or	or	CC
2163	30	34	even	even	RB
2163	30	35	part	part	NN
2163	30	36	of	of	IN
2163	30	37	organization	organization	NN
2163	30	38	names	name	NNS
2163	30	39	(	(	-LRB-
2163	30	40	e.g.	e.g.	RB
2163	30	41	,	,	,
2163	30	42	Max	Max	NNP
2163	30	43	Planck	Planck	NNP
2163	30	44	Institute	Institute	NNP
2163	30	45	)	)	-RRB-
2163	30	46	;	;	:
2163	30	47	in	in	IN
2163	30	48	example	example	NN
2163	30	49	1	1	CD
2163	30	50	a	a	DT
2163	30	51	name	name	NN
2163	30	52	is	be	VBZ
2163	30	53	used	use	VBN
2163	30	54	as	as	IN
2163	30	55	part	part	NN
2163	30	56	of	of	IN
2163	30	57	the	the	DT
2163	30	58	title	title	NN
2163	30	59	;	;	:
2163	30	60			:
2163	30	61	numbers	number	NNS
2163	30	62	—	—	:
2163	30	63	can	can	MD
2163	30	64	act	act	VB
2163	30	65	as	as	IN
2163	30	66	pages	page	NNS
2163	30	67	,	,	,
2163	30	68	years	year	NNS
2163	30	69	,	,	,
2163	30	70	days	day	NNS
2163	30	71	,	,	,
2163	30	72	volume	volume	NN
2163	30	73	numbers	number	NNS
2163	30	74	,	,	,
2163	30	75	or	or	CC
2163	30	76	just	just	RB
2163	30	77	numbers	number	NNS
2163	30	78	within	within	IN
2163	30	79	the	the	DT
2163	30	80	title	title	NN
2163	30	81	;	;	:
2163	30	82			NFP
2163	30	83	locations	location	NNS
2163	30	84	—	—	:
2163	30	85	can	can	MD
2163	30	86	act	act	VB
2163	30	87	as	as	IN
2163	30	88	actual	actual	JJ
2163	30	89	locations	location	NNS
2163	30	90	or	or	CC
2163	30	91	part	part	NN
2163	30	92	of	of	IN
2163	30	93	organization	organization	NN
2163	30	94	names	name	NNS
2163	30	95	(	(	-LRB-
2163	30	96	e.g.	e.g.	RB
2163	30	97	,	,	,
2163	30	98	Univ	Univ	NNP
2163	30	99	.	.	.
2163	31	1	of	of	IN
2163	31	2	Wisconsin	Wisconsin	NNP
2163	31	3	)	)	-RRB-
2163	31	4	To	to	TO
2163	31	5	help	help	VB
2163	31	6	the	the	DT
2163	31	7	chunker	chunker	NN
2163	31	8	in	in	IN
2163	31	9	performing	perform	VBG
2163	31	10	disambiguation	disambiguation	NN
2163	31	11	,	,	,
2163	31	12	one	one	PRP
2163	31	13	can	can	MD
2163	31	14	use	use	VB
2163	31	15	a	a	DT
2163	31	16	series	series	NN
2163	31	17	of	of	IN
2163	31	18	markers	marker	NNS
2163	31	19	,	,	,
2163	31	20	such	such	JJ
2163	31	21	as	as	IN
2163	31	22	,	,	,
2163	31	23	pp	pp	NNP
2163	31	24	.	.	.
2163	32	1	for	for	IN
2163	32	2	pages	page	NNS
2163	32	3	,	,	,
2163	32	4	TR	TR	NNP
2163	32	5	for	for	IN
2163	32	6	technical	technical	JJ
2163	32	7	reports	report	NNS
2163	32	8	,	,	,
2163	32	9	Univ	Univ	NNP
2163	32	10	.	.	.
2163	33	1	or	or	CC
2163	33	2	Institute	Institute	NNP
2163	33	3	for	for	IN
2163	33	4	organization	organization	NN
2163	33	5	.	.	.
2163	34	1	However	however	RB
2163	34	2	,	,	,
2163	34	3	there	there	EX
2163	34	4	are	be	VBP
2163	34	5	cases	case	NNS
2163	34	6	where	where	WRB
2163	34	7	such	such	JJ
2163	34	8	markers	marker	NNS
2163	34	9	help	help	VBP
2163	34	10	in	in	IN
2163	34	11	detecting	detect	VBG
2163	34	12	the	the	DT
2163	34	13	general	general	JJ
2163	34	14	category	category	NN
2163	34	15	of	of	IN
2163	34	16	the	the	DT
2163	34	17	token	token	NN
2163	34	18	,	,	,
2163	34	19	e.g.	e.g.	RB
2163	34	20	,	,	,
2163	34	21	publication	publication	NN
2163	34	22	venue	venue	NN
2163	34	23	,	,	,
2163	34	24	but	but	CC
2163	34	25	a	a	DT
2163	34	26	more	more	RBR
2163	34	27	detailed	detailed	JJ
2163	34	28	disambiguation	disambiguation	NN
2163	34	29	is	be	VBZ
2163	34	30	required	require	VBN
2163	34	31	.	.	.
2163	35	1	For	for	IN
2163	35	2	example	example	NN
2163	35	3	,	,	,
2163	35	4	the	the	DT
2163	35	5	Proc	Proc	NNP
2163	35	6	.	.	.
2163	36	1	marker	marker	NN
2163	36	2	generally	generally	RB
2163	36	3	signals	signal	VBZ
2163	36	4	the	the	DT
2163	36	5	publication	publication	NN
2163	36	6	venue	venue	NN
2163	36	7	of	of	IN
2163	36	8	the	the	DT
2163	36	9	reference	reference	NN
2163	36	10	,	,	,
2163	36	11	without	without	IN
2163	36	12	knowing	know	VBG
2163	36	13	exactly	exactly	RB
2163	36	14	whether	whether	IN
2163	36	15	it	-PRON-	PRP
2163	36	16	represents	represent	VBZ
2163	36	17	a	a	DT
2163	36	18	workshop	workshop	NN
2163	36	19	,	,	,
2163	36	20	conference	conference	NN
2163	36	21	or	or	CC
2163	36	22	even	even	RB
2163	36	23	journal	journal	NNP
2163	36	24	(	(	-LRB-
2163	36	25	as	as	IN
2163	36	26	in	in	IN
2163	36	27	the	the	DT
2163	36	28	case	case	NN
2163	36	29	of	of	IN
2163	36	30	Proc	Proc	NNP
2163	36	31	.	.	.
2163	37	1	Natl	Natl	NNP
2163	37	2	.	.	.
2163	38	1	Acad	Acad	NNS
2163	38	2	.	.	.
2163	39	1	Sci.—Proceedings	Sci.—Proceedings	NNP
2163	39	2	of	of	IN
2163	39	3	the	the	DT
2163	39	4	National	National	NNP
2163	39	5	Academy	Academy	NNP
2163	39	6	of	of	IN
2163	39	7	Sciences	Sciences	NNPS
2163	39	8	)	)	-RRB-
2163	39	9	.	.	.
2163	40	1	The	the	DT
2163	40	2	solution	solution	NN
2163	40	3	we	-PRON-	PRP
2163	40	4	have	have	VBP
2163	40	5	devised	devise	VBN
2163	40	6	was	be	VBD
2163	40	7	built	build	VBN
2163	40	8	to	to	TO
2163	40	9	properly	properly	RB
2163	40	10	handle	handle	VB
2163	40	11	such	such	JJ
2163	40	12	disambiguation	disambiguation	NN
2163	40	13	issues	issue	NNS
2163	40	14	and	and	CC
2163	40	15	the	the	DT
2163	40	16	intrinsic	intrinsic	JJ
2163	40	17	heterogeneous	heterogeneous	JJ
2163	40	18	nature	nature	NN
2163	40	19	of	of	IN
2163	40	20	references	reference	NNS
2163	40	21	.	.	.
2163	41	1	The	the	DT
2163	41	2	features	feature	NNS
2163	41	3	of	of	IN
2163	41	4	the	the	DT
2163	41	5	CRF	CRF	NNP
2163	41	6	chunker	chunker	NN
2163	41	7	model	model	NN
2163	41	8	were	be	VBD
2163	41	9	chosen	choose	VBN
2163	41	10	to	to	TO
2163	41	11	provide	provide	VB
2163	41	12	a	a	DT
2163	41	13	representative	representative	JJ
2163	41	14	discrimination	discrimination	NN
2163	41	15	between	between	IN
2163	41	16	the	the	DT
2163	41	17	different	different	JJ
2163	41	18	fields	field	NNS
2163	41	19	of	of	IN
2163	41	20	the	the	DT
2163	41	21	reference	reference	NN
2163	41	22	string	string	NN
2163	41	23	.	.	.
2163	42	1	Consequently	consequently	RB
2163	42	2	,	,	,
2163	42	3	as	as	IN
2163	42	4	the	the	DT
2163	42	5	experimental	experimental	JJ
2163	42	6	results	result	NNS
2163	42	7	show	show	VBP
2163	42	8	,	,	,
2163	42	9	the	the	DT
2163	42	10	resulting	result	VBG
2163	42	11	chunker	chunker	NN
2163	42	12	has	have	VBZ
2163	42	13	a	a	DT
2163	42	14	superior	superior	JJ
2163	42	15	efficiency	efficiency	NN
2163	42	16	,	,	,
2163	42	17	while	while	IN
2163	42	18	at	at	IN
2163	42	19	the	the	DT
2163	42	20	same	same	JJ
2163	42	21	time	time	NN
2163	42	22	maintaining	maintain	VBG
2163	42	23	an	an	DT
2163	42	24	increased	increase	VBN
2163	42	25	versatility	versatility	NN
2163	42	26	.	.	.
2163	43	1	The	the	DT
2163	43	2	rest	rest	NN
2163	43	3	of	of	IN
2163	43	4	the	the	DT
2163	43	5	paper	paper	NN
2163	43	6	is	be	VBZ
2163	43	7	structured	structure	VBN
2163	43	8	as	as	IN
2163	43	9	follows	follow	VBZ
2163	43	10	:	:	:
2163	43	11	in	in	IN
2163	43	12	section	section	NN
2163	43	13	2	2	CD
2163	43	14	we	-PRON-	PRP
2163	43	15	briefly	briefly	RB
2163	43	16	describe	describe	VBP
2163	43	17	Conditional	Conditional	NNP
2163	43	18	Random	Random	NNP
2163	43	19	Fields	Fields	NNPS
2163	43	20	and	and	CC
2163	43	21	analyze	analyze	VB
2163	43	22	the	the	DT
2163	43	23	existing	exist	VBG
2163	43	24	related	related	JJ
2163	43	25	work	work	NN
2163	43	26	.	.	.
2163	44	1	Section	section	NN
2163	44	2	3	3	CD
2163	44	3	details	detail	VBZ
2163	44	4	the	the	DT
2163	44	5	CRF	CRF	NNP
2163	44	6	-	-	HYPH
2163	44	7	based	base	VBN
2163	44	8	reference	reference	NN
2163	44	9	chunker	chunker	NN
2163	44	10	and	and	CC
2163	44	11	before	before	IN
2163	44	12	concluding	conclude	VBG
2163	44	13	in	in	IN
2163	44	14	section	section	NN
2163	44	15	5	5	CD
2163	44	16	,	,	,
2163	44	17	section	section	NN
2163	44	18	4	4	CD
2163	44	19	presents	present	VBZ
2163	44	20	our	-PRON-	PRP$
2163	44	21	experimental	experimental	JJ
2163	44	22	results	result	NNS
2163	44	23	.	.	.
2163	45	1	2	2	LS
2163	45	2	.	.	.
2163	46	1	BACKGROUND	BACKGROUND	NNP
2163	46	2	2.1	2.1	CD
2163	46	3	Conditional	Conditional	NNP
2163	46	4	Random	Random	NNP
2163	46	5	Fields	Fields	NNPS
2163	46	6	To	to	TO
2163	46	7	have	have	VB
2163	46	8	a	a	DT
2163	46	9	better	well	JJR
2163	46	10	understanding	understanding	NN
2163	46	11	of	of	IN
2163	46	12	the	the	DT
2163	46	13	Machine	Machine	NNP
2163	46	14	Learning	Learning	NNP
2163	46	15	technique	technique	NN
2163	46	16	used	use	VBN
2163	46	17	by	by	IN
2163	46	18	our	-PRON-	PRP$
2163	46	19	solution	solution	NN
2163	46	20	,	,	,
2163	46	21	in	in	IN
2163	46	22	the	the	DT
2163	46	23	following	following	NN
2163	46	24	we	-PRON-	PRP
2163	46	25	give	give	VBP
2163	46	26	a	a	DT
2163	46	27	brief	brief	JJ
2163	46	28	description	description	NN
2163	46	29	of	of	IN
2163	46	30	the	the	DT
2163	46	31	Conditional	Conditional	NNP
2163	46	32	Random	Random	NNP
2163	46	33	Fields	Fields	NNPS
2163	46	34	paradigm	paradigm	NN
2163	46	35	.	.	.
2163	47	1	Figure	figure	NN
2163	47	2	2	2	CD
2163	47	3	.	.	.
2163	48	1	Example	Example	NNP
2163	48	2	Linear	Linear	NNP
2163	48	3	CRF	CRF	NNP
2163	48	4	—	—	:
2163	48	5	Showing	show	VBG
2163	48	6	Dependencies	dependency	NNS
2163	48	7	Between	between	IN
2163	48	8	Features	Features	NNP
2163	48	9	X	X	NNPS
2163	48	10	and	and	CC
2163	48	11	Classes	Classes	NNPS
2163	48	12	Y	Y	NNP
2163	48	13	INFORMATION	information	NN
2163	48	14	TECHNOLOGY	technology	NN
2163	48	15	AND	and	CC
2163	48	16	LIBRARIES	LIBRARIES	NNP
2163	48	17	|	|	NNP
2163	48	18	JUNE	JUNE	NNP
2163	48	19	2012	2012	CD
2163	48	20	9	9	CD
2163	48	21	Conditional	Conditional	NNP
2163	48	22	Random	Random	NNP
2163	48	23	Fields	Fields	NNPS
2163	48	24	(	(	-LRB-
2163	48	25	CRF	CRF	NNP
2163	48	26	)	)	-RRB-
2163	48	27	is	be	VBZ
2163	48	28	a	a	DT
2163	48	29	probabilistic	probabilistic	JJ
2163	48	30	graphical	graphical	JJ
2163	48	31	model	model	NN
2163	48	32	for	for	IN
2163	48	33	classification	classification	NN
2163	48	34	.	.	.
2163	49	1	CRF	CRF	NNP
2163	49	2	,	,	,
2163	49	3	in	in	IN
2163	49	4	general	general	JJ
2163	49	5	,	,	,
2163	49	6	can	can	MD
2163	49	7	represent	represent	VB
2163	49	8	many	many	JJ
2163	49	9	different	different	JJ
2163	49	10	types	type	NNS
2163	49	11	of	of	IN
2163	49	12	graphical	graphical	JJ
2163	49	13	models	model	NNS
2163	49	14	,	,	,
2163	49	15	however	however	RB
2163	49	16	in	in	IN
2163	49	17	the	the	DT
2163	49	18	scope	scope	NN
2163	49	19	of	of	IN
2163	49	20	this	this	DT
2163	49	21	paper	paper	NN
2163	49	22	,	,	,
2163	49	23	we	-PRON-	PRP
2163	49	24	use	use	VBP
2163	49	25	the	the	DT
2163	49	26	so	so	RB
2163	49	27	-	-	HYPH
2163	49	28	called	call	VBN
2163	49	29	linear	linear	NN
2163	49	30	-	-	HYPH
2163	49	31	chain	chain	NN
2163	49	32	CRFs	crf	NNS
2163	49	33	.	.	.
2163	50	1	A	a	DT
2163	50	2	simple	simple	JJ
2163	50	3	example	example	NN
2163	50	4	of	of	IN
2163	50	5	a	a	DT
2163	50	6	linear	linear	JJ
2163	50	7	dependency	dependency	NN
2163	50	8	graph	graph	NN
2163	50	9	is	be	VBZ
2163	50	10	shown	show	VBN
2163	50	11	in	in	IN
2163	50	12	Figure	Figure	NNP
2163	50	13	2	2	CD
2163	50	14	,	,	,
2163	50	15	here	here	RB
2163	50	16	only	only	RB
2163	50	17	the	the	DT
2163	50	18	features	feature	NNS
2163	50	19	X	X	NNS
2163	50	20	of	of	IN
2163	50	21	the	the	DT
2163	50	22	previous	previous	JJ
2163	50	23	item	item	NN
2163	50	24	influences	influence	VBZ
2163	50	25	the	the	DT
2163	50	26	class	class	NN
2163	50	27	of	of	IN
2163	50	28	the	the	DT
2163	50	29	current	current	JJ
2163	50	30	item	item	NN
2163	50	31	Y.	y.	NN
2163	51	1	The	the	DT
2163	51	2	conditional	conditional	JJ
2163	51	3	probability	probability	NN
2163	51	4	is	be	VBZ
2163	51	5	defined	define	VBN
2163	51	6	as	as	IN
2163	51	7	:	:	:
2163	51	8	(	(	-LRB-
2163	51	9	|	|	NNP
2163	51	10	)	)	-RRB-
2163	51	11	(	(	-LRB-
2163	51	12	)	)	-RRB-
2163	51	13	(	(	-LRB-
2163	51	14	∑	∑	.
2163	51	15	(	(	-LRB-
2163	51	16	)	)	-RRB-
2163	51	17	)	)	-RRB-
2163	51	18	where	where	WRB
2163	51	19	(	(	-LRB-
2163	51	20	)	)	-RRB-
2163	51	21	∑	∑	.
2163	51	22	(	(	-LRB-
2163	51	23	)	)	-RRB-
2163	51	24	and	and	CC
2163	51	25	(	(	-LRB-
2163	51	26	)	)	-RRB-
2163	51	27	∑	∑	.
2163	51	28	(	(	-LRB-
2163	51	29	∑	∑	.
2163	51	30	(	(	-LRB-
2163	51	31	)	)	-RRB-
2163	51	32	)	)	-RRB-
2163	51	33	.	.	.
2163	52	1	The	the	DT
2163	52	2	model	model	NN
2163	52	3	is	be	VBZ
2163	52	4	usually	usually	RB
2163	52	5	trained	train	VBN
2163	52	6	by	by	IN
2163	52	7	maximizing	maximize	VBG
2163	52	8	the	the	DT
2163	52	9	log	log	NN
2163	52	10	-	-	HYPH
2163	52	11	likelihood	likelihood	NN
2163	52	12	of	of	IN
2163	52	13	the	the	DT
2163	52	14	training	training	NN
2163	52	15	data	datum	NNS
2163	52	16	by	by	IN
2163	52	17	gradient	gradient	JJ
2163	52	18	methods	method	NNS
2163	52	19	.	.	.
2163	53	1	A	a	DT
2163	53	2	dynamic	dynamic	JJ
2163	53	3	algorithm	algorithm	NN
2163	53	4	is	be	VBZ
2163	53	5	used	use	VBN
2163	53	6	to	to	TO
2163	53	7	compute	compute	VB
2163	53	8	all	all	PDT
2163	53	9	the	the	DT
2163	53	10	required	require	VBN
2163	53	11	probabilities	probability	NNS
2163	53	12	p	p	NNP
2163	53	13	⍬	⍬	NN
2163	53	14	(yi	(yi	NNP
2163	53	15	,	,	,
2163	53	16	yi+1	yi+1	NNP
2163	53	17	)	)	-RRB-
2163	53	18	for	for	IN
2163	53	19	calculating	calculate	VBG
2163	53	20	the	the	DT
2163	53	21	gradient	gradient	NN
2163	53	22	of	of	IN
2163	53	23	the	the	DT
2163	53	24	likelihood	likelihood	NN
2163	53	25	.	.	.
2163	54	1	This	this	DT
2163	54	2	means	mean	VBZ
2163	54	3	that	that	IN
2163	54	4	in	in	IN
2163	54	5	contrast	contrast	NN
2163	54	6	to	to	IN
2163	54	7	traditional	traditional	JJ
2163	54	8	classification	classification	NN
2163	54	9	algorithms	algorithm	NNS
2163	54	10	in	in	IN
2163	54	11	Machine	Machine	NNP
2163	54	12	Learning	Learning	NNP
2163	54	13	(	(	-LRB-
2163	54	14	e.g.	e.g.	RB
2163	54	15	,	,	,
2163	54	16	Support	Support	NNP
2163	54	17	Vector	Vector	NNP
2163	54	18	Machines	Machines	NNP
2163	54	19	9	9	CD
2163	54	20	)	)	-RRB-
2163	54	21	,	,	,
2163	54	22	it	-PRON-	PRP
2163	54	23	not	not	RB
2163	54	24	only	only	RB
2163	54	25	considers	consider	VBZ
2163	54	26	the	the	DT
2163	54	27	attributes	attribute	NNS
2163	54	28	of	of	IN
2163	54	29	the	the	DT
2163	54	30	current	current	JJ
2163	54	31	element	element	NN
2163	54	32	when	when	WRB
2163	54	33	determining	determine	VBG
2163	54	34	the	the	DT
2163	54	35	class	class	NN
2163	54	36	,	,	,
2163	54	37	but	but	CC
2163	54	38	also	also	RB
2163	54	39	attributes	attribute	VBZ
2163	54	40	of	of	IN
2163	54	41	preceding	precede	VBG
2163	54	42	and	and	CC
2163	54	43	succeeding	succeed	VBG
2163	54	44	items	item	NNS
2163	54	45	.	.	.
2163	55	1	This	this	DT
2163	55	2	makes	make	VBZ
2163	55	3	it	-PRON-	PRP
2163	55	4	ideal	ideal	JJ
2163	55	5	for	for	IN
2163	55	6	tagging	tag	VBG
2163	55	7	sequences	sequence	NNS
2163	55	8	,	,	,
2163	55	9	such	such	JJ
2163	55	10	as	as	IN
2163	55	11	chunking	chunk	VBG
2163	55	12	of	of	IN
2163	55	13	parts	part	NNS
2163	55	14	of	of	IN
2163	55	15	speech	speech	NN
2163	55	16	or	or	CC
2163	55	17	parts	part	NNS
2163	55	18	of	of	IN
2163	55	19	references	reference	NNS
2163	55	20	,	,	,
2163	55	21	which	which	WDT
2163	55	22	is	be	VBZ
2163	55	23	what	what	WP
2163	55	24	we	-PRON-	PRP
2163	55	25	require	require	VBP
2163	55	26	for	for	IN
2163	55	27	our	-PRON-	PRP$
2163	55	28	chunking	chunk	VBG
2163	55	29	task	task	NN
2163	55	30	.	.	.
2163	56	1	2.2	2.2	CD
2163	56	2	Related	related	JJ
2163	56	3	Work	work	NN
2163	56	4	In	in	IN
2163	56	5	recent	recent	JJ
2163	56	6	years	year	NNS
2163	56	7	,	,	,
2163	56	8	extensive	extensive	JJ
2163	56	9	research	research	NN
2163	56	10	has	have	VBZ
2163	56	11	been	be	VBN
2163	56	12	performed	perform	VBN
2163	56	13	in	in	IN
2163	56	14	the	the	DT
2163	56	15	area	area	NN
2163	56	16	of	of	IN
2163	56	17	automatic	automatic	JJ
2163	56	18	metadata	metadata	NN
2163	56	19	extraction	extraction	NN
2163	56	20	from	from	IN
2163	56	21	scientific	scientific	JJ
2163	56	22	publications	publication	NNS
2163	56	23	.	.	.
2163	57	1	Most	Most	JJS
2163	57	2	of	of	IN
2163	57	3	the	the	DT
2163	57	4	approaches	approaches	NNP
2163	57	5	focus	focus	VBP
2163	57	6	on	on	IN
2163	57	7	one	one	CD
2163	57	8	of	of	IN
2163	57	9	the	the	DT
2163	57	10	two	two	CD
2163	57	11	main	main	JJ
2163	57	12	metadata	metadata	NN
2163	57	13	components	component	NNS
2163	57	14	,	,	,
2163	57	15	i.e.	i.e.	FW
2163	57	16	,	,	,
2163	57	17	on	on	IN
2163	57	18	the	the	DT
2163	57	19	heading	heading	NN
2163	57	20	/	/	SYM
2163	57	21	bibliographic	bibliographic	JJ
2163	57	22	metadata	metadata	NN
2163	57	23	or	or	CC
2163	57	24	on	on	IN
2163	57	25	the	the	DT
2163	57	26	reference	reference	NN
2163	57	27	metadata	metadata	NN
2163	57	28	,	,	,
2163	57	29	but	but	CC
2163	57	30	there	there	EX
2163	57	31	are	be	VBP
2163	57	32	also	also	RB
2163	57	33	cases	case	NNS
2163	57	34	when	when	WRB
2163	57	35	the	the	DT
2163	57	36	entire	entire	JJ
2163	57	37	set	set	NN
2163	57	38	is	be	VBZ
2163	57	39	targeted	target	VBN
2163	57	40	.	.	.
2163	58	1	As	as	IN
2163	58	2	this	this	DT
2163	58	3	paper	paper	NN
2163	58	4	focuses	focus	VBZ
2163	58	5	only	only	RB
2163	58	6	on	on	IN
2163	58	7	the	the	DT
2163	58	8	second	second	JJ
2163	58	9	component	component	NN
2163	58	10	,	,	,
2163	58	11	within	within	IN
2163	58	12	this	this	DT
2163	58	13	section	section	NN
2163	58	14	we	-PRON-	PRP
2163	58	15	present	present	VBP
2163	58	16	and	and	CC
2163	58	17	discuss	discuss	VB
2163	58	18	those	those	DT
2163	58	19	applications	application	NNS
2163	58	20	that	that	WDT
2163	58	21	deal	deal	VBP
2163	58	22	strictly	strictly	RB
2163	58	23	with	with	IN
2163	58	24	reference	reference	NN
2163	58	25	chunking	chunk	VBG
2163	58	26	.	.	.
2163	59	1	The	the	DT
2163	59	2	ParsCit	ParsCit	NNP
2163	59	3	framework	framework	NN
2163	59	4	is	be	VBZ
2163	59	5	the	the	DT
2163	59	6	closest	close	JJS
2163	59	7	technique	technique	NN
2163	59	8	mapping	mapping	NN
2163	59	9	to	to	IN
2163	59	10	our	-PRON-	PRP$
2163	59	11	goals	goal	NNS
2163	59	12	and	and	CC
2163	59	13	methodology	methodology	NN
2163	59	14	.	.	.
2163	60	1	10	10	CD
2163	60	2	ParsCit	ParsCit	NNP
2163	60	3	is	be	VBZ
2163	60	4	an	an	DT
2163	60	5	open	open	JJ
2163	60	6	-	-	HYPH
2163	60	7	source	source	NN
2163	60	8	reference	reference	NN
2163	60	9	-	-	HYPH
2163	60	10	parsing	parse	VBG
2163	60	11	package	package	NN
2163	60	12	.	.	.
2163	61	1	While	while	IN
2163	61	2	its	-PRON-	PRP$
2163	61	3	first	first	JJ
2163	61	4	version	version	NN
2163	61	5	used	use	VBD
2163	61	6	a	a	DT
2163	61	7	Maximum	Maximum	NNP
2163	61	8	Entropy	Entropy	NNP
2163	61	9	model	model	NN
2163	61	10	to	to	TO
2163	61	11	perform	perform	VB
2163	61	12	reference	reference	NN
2163	61	13	chunking	chunk	VBG
2163	61	14	,	,	,
2163	61	15	11	11	CD
2163	61	16	currently	currently	RB
2163	61	17	,	,	,
2163	61	18	inspired	inspire	VBN
2163	61	19	by	by	IN
2163	61	20	the	the	DT
2163	61	21	work	work	NN
2163	61	22	of	of	IN
2163	61	23	Peng	Peng	NNP
2163	61	24	et	et	NNP
2163	61	25	al	al	NNP
2163	61	26	.	.	.
2163	62	1	,	,	,
2163	62	2	12	12	CD
2163	62	3	it	-PRON-	PRP
2163	62	4	uses	use	VBZ
2163	62	5	a	a	DT
2163	62	6	trained	train	VBN
2163	62	7	CRF	CRF	NNP
2163	62	8	model	model	NN
2163	62	9	for	for	IN
2163	62	10	label	label	NNP
2163	62	11	sequencing	sequencing	NN
2163	62	12	.	.	.
2163	63	1	The	the	DT
2163	63	2	model	model	NN
2163	63	3	was	be	VBD
2163	63	4	obtained	obtain	VBN
2163	63	5	based	base	VBN
2163	63	6	on	on	IN
2163	63	7	a	a	DT
2163	63	8	set	set	NN
2163	63	9	of	of	IN
2163	63	10	twenty	twenty	CD
2163	63	11	-	-	HYPH
2163	63	12	three	three	CD
2163	63	13	token	token	NN
2163	63	14	-	-	HYPH
2163	63	15	oriented	orient	VBN
2163	63	16	features	feature	NNS
2163	63	17	tailored	tailor	VBN
2163	63	18	towards	towards	IN
2163	63	19	correcting	correct	VBG
2163	63	20	the	the	DT
2163	63	21	errors	error	NNS
2163	63	22	that	that	WDT
2163	63	23	Peng	Peng	NNP
2163	63	24	's	's	POS
2163	63	25	CRF	CRF	NNP
2163	63	26	model	model	NN
2163	63	27	produced	produce	VBD
2163	63	28	.	.	.
2163	64	1	Our	-PRON-	PRP$
2163	64	2	CRF	CRF	NNP
2163	64	3	chunker	chunker	NN
2163	64	4	builds	build	VBZ
2163	64	5	on	on	IN
2163	64	6	the	the	DT
2163	64	7	work	work	NN
2163	64	8	of	of	IN
2163	64	9	ParsCit	ParsCit	NNP
2163	64	10	.	.	.
2163	65	1	However	however	RB
2163	65	2	,	,	,
2163	65	3	as	as	IN
2163	65	4	we	-PRON-	PRP
2163	65	5	aimed	aim	VBD
2163	65	6	at	at	IN
2163	65	7	improving	improve	VBG
2163	65	8	the	the	DT
2163	65	9	chunking	chunk	VBG
2163	65	10	performance	performance	NN
2163	65	11	,	,	,
2163	65	12	we	-PRON-	PRP
2163	65	13	altered	alter	VBD
2163	65	14	some	some	DT
2163	65	15	of	of	IN
2163	65	16	the	the	DT
2163	65	17	existing	exist	VBG
2163	65	18	features	feature	NNS
2163	65	19	and	and	CC
2163	65	20	introduced	introduce	VBD
2163	65	21	additional	additional	JJ
2163	65	22	ones	one	NNS
2163	65	23	.	.	.
2163	66	1	Moreover	moreover	RB
2163	66	2	,	,	,
2163	66	3	we	-PRON-	PRP
2163	66	4	have	have	VBP
2163	66	5	compiled	compile	VBN
2163	66	6	significantly	significantly	RB
2163	66	7	larger	large	JJR
2163	66	8	gazetteers	gazetteer	NNS
2163	66	9	required	require	VBN
2163	66	10	for	for	IN
2163	66	11	detecting	detect	VBG
2163	66	12	different	different	JJ
2163	66	13	aspects	aspect	NNS
2163	66	14	,	,	,
2163	66	15	such	such	JJ
2163	66	16	as	as	IN
2163	66	17	names	name	NNS
2163	66	18	,	,	,
2163	66	19	places	place	NNS
2163	66	20	,	,	,
2163	66	21	organizations	organization	NNS
2163	66	22	,	,	,
2163	66	23	journals	journal	NNS
2163	66	24	,	,	,
2163	66	25	or	or	CC
2163	66	26	publishers	publisher	NNS
2163	66	27	.	.	.
2163	67	1	One	one	CD
2163	67	2	of	of	IN
2163	67	3	the	the	DT
2163	67	4	first	first	JJ
2163	67	5	attempts	attempt	NNS
2163	67	6	to	to	TO
2163	67	7	extract	extract	VB
2163	67	8	and	and	CC
2163	67	9	index	index	NN
2163	67	10	reference	reference	NN
2163	67	11	information	information	NN
2163	67	12	led	lead	VBD
2163	67	13	to	to	IN
2163	67	14	the	the	DT
2163	67	15	currently	currently	RB
2163	67	16	well-	well-	XX
2163	67	17	known	know	VBN
2163	67	18	system	system	NN
2163	67	19	,	,	,
2163	67	20	CiteSeer	CiteSeer	NNP
2163	67	21	.	.	.
2163	68	1	13	13	CD
2163	68	2	Around	around	IN
2163	68	3	the	the	DT
2163	68	4	same	same	JJ
2163	68	5	period	period	NN
2163	68	6	,	,	,
2163	68	7	Seymore	Seymore	NNP
2163	68	8	et	et	NNP
2163	68	9	al	al	NNP
2163	68	10	.	.	.
2163	69	1	developed	develop	VBD
2163	69	2	one	one	CD
2163	69	3	of	of	IN
2163	69	4	the	the	DT
2163	69	5	first	first	JJ
2163	69	6	reference	reference	NN
2163	69	7	chunking	chunk	VBG
2163	69	8	approaches	approach	NNS
2163	69	9	that	that	WDT
2163	69	10	used	use	VBD
2163	69	11	Machine	Machine	NNP
2163	69	12	Learning	Learning	NNP
2163	69	13	techniques	technique	NNS
2163	69	14	.	.	.
2163	70	1	14	14	CD
2163	70	2	The	the	DT
2163	70	3	authors	author	NNS
2163	70	4	trained	train	VBD
2163	70	5	a	a	DT
2163	70	6	Hidden	Hidden	NNP
2163	70	7	Markov	Markov	NNP
2163	70	8	Model	Model	NNP
2163	70	9	(	(	-LRB-
2163	70	10	HMM	HMM	NNP
2163	70	11	)	)	-RRB-
2163	70	12	to	to	TO
2163	70	13	build	build	VB
2163	70	14	a	a	DT
2163	70	15	reference	reference	NN
2163	70	16	sequence	sequence	NN
2163	70	17	labeler	labeler	NN
2163	70	18	using	use	VBG
2163	70	19	internal	internal	JJ
2163	70	20	states	state	NNS
2163	70	21	for	for	IN
2163	70	22	different	different	JJ
2163	70	23	parts	part	NNS
2163	70	24	of	of	IN
2163	70	25	the	the	DT
2163	70	26	fields	field	NNS
2163	70	27	.	.	.
2163	71	1	As	as	IN
2163	71	2	it	-PRON-	PRP
2163	71	3	represented	represent	VBD
2163	71	4	pioneering	pioneer	VBG
2163	71	5	work	work	NN
2163	71	6	,	,	,
2163	71	7	it	-PRON-	PRP
2163	71	8	also	also	RB
2163	71	9	resulted	result	VBD
2163	71	10	in	in	IN
2163	71	11	the	the	DT
2163	71	12	first	first	JJ
2163	71	13	gold	gold	JJ
2163	71	14	standard	standard	JJ
2163	71	15	set	set	NN
2163	71	16	,	,	,
2163	71	17	the	the	DT
2163	71	18	CORA	CORA	NNP
2163	71	19	dataset	dataset	NN
2163	71	20	.	.	.
2163	72	1	At	at	IN
2163	72	2	a	a	DT
2163	72	3	later	later	JJ
2163	72	4	stage	stage	NN
2163	72	5	,	,	,
2163	72	6	the	the	DT
2163	72	7	same	same	JJ
2163	72	8	group	group	NN
2163	72	9	applied	apply	VBD
2163	72	10	CRF	CRF	NNP
2163	72	11	for	for	IN
2163	72	12	the	the	DT
2163	72	13	first	first	JJ
2163	72	14	time	time	NN
2163	72	15	to	to	TO
2163	72	16	perform	perform	VB
2163	72	17	reference	reference	NN
2163	72	18	chunking	chunking	NN
2163	72	19	,	,	,
2163	72	20	which	which	WDT
2163	72	21	later	later	RB
2163	72	22	inspired	inspire	VBD
2163	72	23	ParsCit	ParsCit	NNP
2163	72	24	.	.	.
2163	73	1	15	15	CD
2163	73	2	REFERENCE	reference	NN
2163	73	3	INFORMATION	INFORMATION	VBN
2163	73	4	EXTRACTION	extraction	NN
2163	73	5	AND	and	CC
2163	73	6	PROCESSING	processing	NN
2163	73	7	|GROZA	|groza	NN
2163	73	8	,	,	,
2163	73	9	GRIMNES	GRIMNES	NNP
2163	73	10	,	,	,
2163	73	11	AND	and	CC
2163	73	12	HANDSCHUH	handschuh	NN
2163	73	13	10	10	CD
2163	73	14	In	in	IN
2163	73	15	the	the	DT
2163	73	16	same	same	JJ
2163	73	17	learning	learning	NN
2163	73	18	-	-	HYPH
2163	73	19	driven	drive	VBN
2163	73	20	category	category	NN
2163	73	21	is	be	VBZ
2163	73	22	the	the	DT
2163	73	23	work	work	NN
2163	73	24	of	of	IN
2163	73	25	Han	Han	NNP
2163	73	26	et	et	NNP
2163	73	27	al	al	NNP
2163	73	28	.	.	.
2163	74	1	16	16	CD
2163	74	2	The	the	DT
2163	74	3	authors	author	NNS
2163	74	4	proposed	propose	VBD
2163	74	5	an	an	DT
2163	74	6	effective	effective	JJ
2163	74	7	word	word	NN
2163	74	8	clustering	cluster	VBG
2163	74	9	approach	approach	NN
2163	74	10	with	with	IN
2163	74	11	the	the	DT
2163	74	12	goal	goal	NN
2163	74	13	of	of	IN
2163	74	14	reducing	reduce	VBG
2163	74	15	feature	feature	NN
2163	74	16	dimensionality	dimensionality	NN
2163	74	17	when	when	WRB
2163	74	18	compared	compare	VBN
2163	74	19	to	to	IN
2163	74	20	HMM	HMM	NNP
2163	74	21	,	,	,
2163	74	22	while	while	IN
2163	74	23	at	at	IN
2163	74	24	the	the	DT
2163	74	25	same	same	JJ
2163	74	26	time	time	NN
2163	74	27	improving	improve	VBG
2163	74	28	the	the	DT
2163	74	29	overall	overall	JJ
2163	74	30	chunking	chunk	VBG
2163	74	31	performance	performance	NN
2163	74	32	.	.	.
2163	75	1	The	the	DT
2163	75	2	resultant	resultant	JJ
2163	75	3	domain	domain	NN
2163	75	4	,	,	,
2163	75	5	rule	rule	NN
2163	75	6	-	-	HYPH
2163	75	7	based	base	VBN
2163	75	8	word	word	NN
2163	75	9	clustering	clustering	NN
2163	75	10	method	method	NN
2163	75	11	for	for	IN
2163	75	12	cluster	cluster	NN
2163	75	13	feature	feature	NN
2163	75	14	representation	representation	NN
2163	75	15	used	use	VBD
2163	75	16	clusters	cluster	NNS
2163	75	17	formed	form	VBN
2163	75	18	from	from	IN
2163	75	19	various	various	JJ
2163	75	20	domain	domain	NN
2163	75	21	databases	database	NNS
2163	75	22	and	and	CC
2163	75	23	word	word	NN
2163	75	24	orthographic	orthographic	JJ
2163	75	25	properties	property	NNS
2163	75	26	.	.	.
2163	76	1	Consequently	consequently	RB
2163	76	2	,	,	,
2163	76	3	they	-PRON-	PRP
2163	76	4	achieved	achieve	VBD
2163	76	5	an	an	DT
2163	76	6	8.5	8.5	CD
2163	76	7	percent	percent	NN
2163	76	8	improvement	improvement	NN
2163	76	9	on	on	IN
2163	76	10	the	the	DT
2163	76	11	overall	overall	JJ
2163	76	12	accuracy	accuracy	NN
2163	76	13	of	of	IN
2163	76	14	reference	reference	NN
2163	76	15	fields	field	VBZ
2163	76	16	classification	classification	NN
2163	76	17	combined	combine	VBN
2163	76	18	with	with	IN
2163	76	19	a	a	DT
2163	76	20	significant	significant	JJ
2163	76	21	dimensionality	dimensionality	NN
2163	76	22	reduction	reduction	NN
2163	76	23	.	.	.
2163	77	1	FLUX	FLUX	NNP
2163	77	2	-	-	HYPH
2163	77	3	CIM	CIM	NNP
2163	77	4	17	17	CD
2163	77	5	is	be	VBZ
2163	77	6	the	the	DT
2163	77	7	only	only	JJ
2163	77	8	unsupervised	unsupervise	VBN
2163	77	9	18	18	CD
2163	77	10	approach	approach	NN
2163	77	11	that	that	WDT
2163	77	12	targets	target	VBZ
2163	77	13	reference	reference	NN
2163	77	14	chunking	chunk	VBG
2163	77	15	.	.	.
2163	78	1	The	the	DT
2163	78	2	system	system	NN
2163	78	3	uses	use	VBZ
2163	78	4	automatically	automatically	RB
2163	78	5	constructed	construct	VBN
2163	78	6	knowledge	knowledge	NN
2163	78	7	bases	basis	NNS
2163	78	8	from	from	IN
2163	78	9	an	an	DT
2163	78	10	existing	exist	VBG
2163	78	11	set	set	NN
2163	78	12	of	of	IN
2163	78	13	sample	sample	NN
2163	78	14	references	reference	NNS
2163	78	15	for	for	IN
2163	78	16	recognizing	recognize	VBG
2163	78	17	the	the	DT
2163	78	18	component	component	NN
2163	78	19	fields	field	NNS
2163	78	20	of	of	IN
2163	78	21	a	a	DT
2163	78	22	reference	reference	NN
2163	78	23	.	.	.
2163	79	1	The	the	DT
2163	79	2	chunking	chunking	NN
2163	79	3	process	process	NN
2163	79	4	features	feature	VBZ
2163	79	5	two	two	CD
2163	79	6	steps	step	NNS
2163	79	7	:	:	:
2163	79	8			:
2163	79	9	a	a	DT
2163	79	10	probability	probability	NN
2163	79	11	estimation	estimation	NN
2163	79	12	of	of	IN
2163	79	13	a	a	DT
2163	79	14	given	give	VBN
2163	79	15	term	term	NN
2163	79	16	within	within	IN
2163	79	17	a	a	DT
2163	79	18	reference	reference	NN
2163	79	19	which	which	WDT
2163	79	20	is	be	VBZ
2163	79	21	a	a	DT
2163	79	22	value	value	NN
2163	79	23	for	for	IN
2163	79	24	a	a	DT
2163	79	25	given	give	VBN
2163	79	26	reference	reference	NN
2163	79	27	field	field	NN
2163	79	28	based	base	VBN
2163	79	29	on	on	IN
2163	79	30	the	the	DT
2163	79	31	information	information	NN
2163	79	32	encoded	encode	VBN
2163	79	33	in	in	IN
2163	79	34	their	-PRON-	PRP$
2163	79	35	knowledge	knowledge	NN
2163	79	36	bases	basis	NNS
2163	79	37	,	,	,
2163	79	38	and	and	CC
2163	79	39			NFP
2163	79	40	the	the	DT
2163	79	41	use	use	NN
2163	79	42	of	of	IN
2163	79	43	generic	generic	JJ
2163	79	44	structural	structural	JJ
2163	79	45	properties	property	NNS
2163	79	46	of	of	IN
2163	79	47	references	reference	NNS
2163	79	48	.	.	.
2163	80	1	Similarly	similarly	RB
2163	80	2	to	to	IN
2163	80	3	Seymore	Seymore	NNP
2163	80	4	et	et	NNP
2163	80	5	al	al	NNP
2163	80	6	.	.	NNP
2163	80	7	,	,	,
2163	80	8	19	19	CD
2163	80	9	the	the	DT
2163	80	10	authors	author	NNS
2163	80	11	have	have	VBP
2163	80	12	also	also	RB
2163	80	13	created	create	VBN
2163	80	14	two	two	CD
2163	80	15	datasets	dataset	NNS
2163	80	16	(	(	-LRB-
2163	80	17	specifically	specifically	RB
2163	80	18	for	for	IN
2163	80	19	the	the	DT
2163	80	20	computer	computer	NN
2163	80	21	science	science	NN
2163	80	22	and	and	CC
2163	80	23	health	health	NN
2163	80	24	science	science	NN
2163	80	25	areas	area	NNS
2163	80	26	)	)	-RRB-
2163	80	27	to	to	TO
2163	80	28	be	be	VB
2163	80	29	used	use	VBN
2163	80	30	for	for	IN
2163	80	31	comparing	compare	VBG
2163	80	32	the	the	DT
2163	80	33	achieved	achieve	VBN
2163	80	34	accuracies	accuracy	NNS
2163	80	35	.	.	.
2163	81	1	A	a	DT
2163	81	2	completely	completely	RB
2163	81	3	different	different	JJ
2163	81	4	,	,	,
2163	81	5	and	and	CC
2163	81	6	novel	novel	NN
2163	81	7	,	,	,
2163	81	8	direction	direction	NN
2163	81	9	was	be	VBD
2163	81	10	developed	develop	VBN
2163	81	11	by	by	IN
2163	81	12	Poon	Poon	NNP
2163	81	13	and	and	CC
2163	81	14	Domingos	Domingos	NNP
2163	81	15	.	.	.
2163	82	1	20	20	CD
2163	82	2	Unlike	unlike	IN
2163	82	3	all	all	PDT
2163	82	4	the	the	DT
2163	82	5	other	other	JJ
2163	82	6	approaches	approach	NNS
2163	82	7	,	,	,
2163	82	8	they	-PRON-	PRP
2163	82	9	propose	propose	VBP
2163	82	10	a	a	DT
2163	82	11	solution	solution	NN
2163	82	12	where	where	WRB
2163	82	13	the	the	DT
2163	82	14	segmentation	segmentation	NN
2163	82	15	(	(	-LRB-
2163	82	16	chunking	chunking	NN
2163	82	17	)	)	-RRB-
2163	82	18	of	of	IN
2163	82	19	the	the	DT
2163	82	20	reference	reference	NN
2163	82	21	fields	field	NNS
2163	82	22	is	be	VBZ
2163	82	23	performed	perform	VBN
2163	82	24	together	together	RB
2163	82	25	with	with	IN
2163	82	26	the	the	DT
2163	82	27	entity	entity	NN
2163	82	28	resolution	resolution	NN
2163	82	29	in	in	IN
2163	82	30	a	a	DT
2163	82	31	single	single	JJ
2163	82	32	integrated	integrated	JJ
2163	82	33	inference	inference	NN
2163	82	34	process	process	NN
2163	82	35	.	.	.
2163	83	1	They	-PRON-	PRP
2163	83	2	,	,	,
2163	83	3	thus	thus	RB
2163	83	4	,	,	,
2163	83	5	help	help	NN
2163	83	6	in	in	IN
2163	83	7	disambiguating	disambiguate	VBG
2163	83	8	the	the	DT
2163	83	9	boundaries	boundary	NNS
2163	83	10	of	of	IN
2163	83	11	less	less	RBR
2163	83	12	-	-	HYPH
2163	83	13	clear	clear	RB
2163	83	14	chunked	chunk	VBN
2163	83	15	fields	field	NNS
2163	83	16	,	,	,
2163	83	17	using	use	VBG
2163	83	18	the	the	DT
2163	83	19	already	already	RB
2163	83	20	well	well	RB
2163	83	21	-	-	HYPH
2163	83	22	segmented	segment	VBN
2163	83	23	ones	one	NNS
2163	83	24	.	.	.
2163	84	1	Although	although	IN
2163	84	2	the	the	DT
2163	84	3	results	result	NNS
2163	84	4	achieved	achieve	VBN
2163	84	5	are	be	VBP
2163	84	6	similar	similar	JJ
2163	84	7	to	to	IN
2163	84	8	,	,	,
2163	84	9	and	and	CC
2163	84	10	even	even	RB
2163	84	11	better	well	JJR
2163	84	12	than	than	IN
2163	84	13	some	some	DT
2163	84	14	of	of	IN
2163	84	15	,	,	,
2163	84	16	the	the	DT
2163	84	17	above	above	RB
2163	84	18	-	-	HYPH
2163	84	19	mentioned	mention	VBN
2163	84	20	approaches	approach	NNS
2163	84	21	,	,	,
2163	84	22	this	this	DT
2163	84	23	is	be	VBZ
2163	84	24	suboptimal	suboptimal	JJ
2163	84	25	from	from	IN
2163	84	26	the	the	DT
2163	84	27	computational	computational	JJ
2163	84	28	perspective	perspective	NN
2163	84	29	:	:	:
2163	84	30	the	the	DT
2163	84	31	chunking	chunking	NN
2163	84	32	/	/	SYM
2163	84	33	resolution	resolution	NN
2163	84	34	time	time	NN
2163	84	35	reported	report	VBN
2163	84	36	by	by	IN
2163	84	37	the	the	DT
2163	84	38	authors	author	NNS
2163	84	39	measured	measure	VBN
2163	84	40	around	around	RB
2163	84	41	thirty	thirty	CD
2163	84	42	minutes	minute	NNS
2163	84	43	.	.	.
2163	85	1	In	in	IN
2163	85	2	addition	addition	NN
2163	85	3	to	to	IN
2163	85	4	the	the	DT
2163	85	5	previously	previously	RB
2163	85	6	described	describe	VBN
2163	85	7	works	work	NNS
2163	85	8	,	,	,
2163	85	9	which	which	WDT
2163	85	10	were	be	VBD
2163	85	11	specifically	specifically	RB
2163	85	12	tailored	tailor	VBN
2163	85	13	for	for	IN
2163	85	14	bibliographic	bibliographic	JJ
2163	85	15	metadata	metadata	NN
2163	85	16	extraction	extraction	NN
2163	85	17	,	,	,
2163	85	18	there	there	EX
2163	85	19	are	be	VBP
2163	85	20	a	a	DT
2163	85	21	series	series	NN
2163	85	22	of	of	IN
2163	85	23	other	other	JJ
2163	85	24	approaches	approach	NNS
2163	85	25	that	that	WDT
2163	85	26	could	could	MD
2163	85	27	be	be	VB
2163	85	28	used	use	VBN
2163	85	29	for	for	IN
2163	85	30	the	the	DT
2163	85	31	same	same	JJ
2163	85	32	purpose	purpose	NN
2163	85	33	.	.	.
2163	86	1	For	for	IN
2163	86	2	example	example	NN
2163	86	3	,	,	,
2163	86	4	Cesario	Cesario	NNP
2163	86	5	et	et	NNP
2163	86	6	al	al	NNP
2163	86	7	.	.	.
2163	87	1	propose	propose	VB
2163	87	2	an	an	DT
2163	87	3	innovative	innovative	JJ
2163	87	4	recursive	recursive	JJ
2163	87	5	boosting	boosting	NN
2163	87	6	strategy	strategy	NN
2163	87	7	,	,	,
2163	87	8	with	with	IN
2163	87	9	progressive	progressive	JJ
2163	87	10	classification	classification	NN
2163	87	11	,	,	,
2163	87	12	to	to	TO
2163	87	13	reconcile	reconcile	VB
2163	87	14	textual	textual	JJ
2163	87	15	elements	element	NNS
2163	87	16	to	to	IN
2163	87	17	an	an	DT
2163	87	18	existing	exist	VBG
2163	87	19	attribute	attribute	NN
2163	87	20	schema	schema	NN
2163	87	21	.	.	.
2163	88	1	21	21	CD
2163	88	2	In	in	IN
2163	88	3	the	the	DT
2163	88	4	case	case	NN
2163	88	5	of	of	IN
2163	88	6	bibliographic	bibliographic	JJ
2163	88	7	metadata	metadata	NN
2163	88	8	segmentation	segmentation	NN
2163	88	9	,	,	,
2163	88	10	the	the	DT
2163	88	11	metadata	metadata	NN
2163	88	12	fields	field	NNS
2163	88	13	would	would	MD
2163	88	14	correspond	correspond	VB
2163	88	15	to	to	IN
2163	88	16	the	the	DT
2163	88	17	textual	textual	JJ
2163	88	18	elements	element	NNS
2163	88	19	,	,	,
2163	88	20	while	while	IN
2163	88	21	an	an	DT
2163	88	22	ontology	ontology	NN
2163	88	23	describing	describe	VBG
2163	88	24	them	-PRON-	PRP
2163	88	25	(	(	-LRB-
2163	88	26	e.g.	e.g.	RB
2163	88	27	,	,	,
2163	88	28	DublinCore	DublinCore	NNP
2163	88	29	22	22	CD
2163	88	30	or	or	CC
2163	88	31	SWRC	SWRC	NNP
2163	88	32	23	23	CD
2163	88	33	)	)	-RRB-
2163	88	34	would	would	MD
2163	88	35	have	have	VB
2163	88	36	the	the	DT
2163	88	37	schema	schema	NNP
2163	88	38	role	role	NN
2163	88	39	.	.	.
2163	89	1	The	the	DT
2163	89	2	authors	author	NNS
2163	89	3	even	even	RB
2163	89	4	describe	describe	VBP
2163	89	5	an	an	DT
2163	89	6	evaluation	evaluation	NN
2163	89	7	of	of	IN
2163	89	8	the	the	DT
2163	89	9	method	method	NN
2163	89	10	using	use	VBG
2163	89	11	the	the	DT
2163	89	12	DBLP	DBLP	NNP
2163	89	13	citation	citation	NN
2163	89	14	dataset	dataset	NN
2163	89	15	,	,	,
2163	89	16	however	however	RB
2163	89	17	,	,	,
2163	89	18	without	without	IN
2163	89	19	giving	give	VBG
2163	89	20	precise	precise	JJ
2163	89	21	details	detail	NNS
2163	89	22	on	on	IN
2163	89	23	the	the	DT
2163	89	24	fields	field	NNS
2163	89	25	considered	consider	VBN
2163	89	26	for	for	IN
2163	89	27	segmentation	segmentation	NN
2163	89	28	.	.	.
2163	90	1	Some	some	DT
2163	90	2	other	other	JJ
2163	90	3	approaches	approach	NNS
2163	90	4	include	include	VBP
2163	90	5	,	,	,
2163	90	6	in	in	IN
2163	90	7	general	general	JJ
2163	90	8	,	,	,
2163	90	9	any	any	DT
2163	90	10	sequence	sequence	NN
2163	90	11	labeling	labeling	NN
2163	90	12	techniques	technique	NNS
2163	90	13	,	,	,
2163	90	14	e.g.	e.g.	RB
2163	90	15	,	,	,
2163	90	16	SLF	SLF	NNP
2163	90	17	,	,	,
2163	90	18	24	24	CD
2163	90	19	named	name	VBN
2163	90	20	entity	entity	NN
2163	90	21	recognition	recognition	NN
2163	90	22	techniques	technique	NNS
2163	90	23	,	,	,
2163	90	24	25	25	CD
2163	90	25	or	or	CC
2163	90	26	even	even	RB
2163	90	27	Field	Field	NNP
2163	90	28	Association	Association	NNP
2163	90	29	(	(	-LRB-
2163	90	30	FA	FA	NNP
2163	90	31	)	)	-RRB-
2163	90	32	terms	term	NNS
2163	90	33	extraction	extraction	NN
2163	90	34	,	,	,
2163	90	35	26	26	CD
2163	90	36	the	the	DT
2163	90	37	latter	latter	JJ
2163	90	38	working	work	VBG
2163	90	39	on	on	IN
2163	90	40	bibliographic	bibliographic	JJ
2163	90	41	metadata	metadata	NN
2163	90	42	fields	field	NNS
2163	90	43	in	in	IN
2163	90	44	a	a	DT
2163	90	45	quasi	quasi	JJ
2163	90	46	-	-	JJ
2163	90	47	similar	similar	JJ
2163	90	48	manner	manner	NN
2163	90	49	as	as	IN
2163	90	50	the	the	DT
2163	90	51	recursive	recursive	JJ
2163	90	52	boosting	boosting	NN
2163	90	53	strategy	strategy	NN
2163	90	54	.	.	.
2163	91	1	In	in	IN
2163	91	2	conclusion	conclusion	NN
2163	91	3	,	,	,
2163	91	4	it	-PRON-	PRP
2163	91	5	is	be	VBZ
2163	91	6	worth	worth	JJ
2163	91	7	mentioning	mention	VBG
2163	91	8	that	that	IN
2163	91	9	retrieving	retrieve	VBG
2163	91	10	citation	citation	NNP
2163	91	11	contexts	contexts	NNP
2163	91	12	is	be	VBZ
2163	91	13	an	an	DT
2163	91	14	interesting	interesting	JJ
2163	91	15	research	research	NN
2163	91	16	area	area	NN
2163	91	17	especially	especially	RB
2163	91	18	in	in	IN
2163	91	19	the	the	DT
2163	91	20	context	context	NN
2163	91	21	of	of	IN
2163	91	22	digital	digital	JJ
2163	91	23	libraries	library	NNS
2163	91	24	.	.	.
2163	92	1	Our	-PRON-	PRP$
2163	92	2	current	current	JJ
2163	92	3	work	work	NN
2163	92	4	does	do	VBZ
2163	92	5	not	not	RB
2163	92	6	feature	feature	VB
2163	92	7	this	this	DT
2163	92	8	aspect	aspect	NN
2163	92	9	,	,	,
2163	92	10	but	but	CC
2163	92	11	we	-PRON-	PRP
2163	92	12	regard	regard	VBP
2163	92	13	it	-PRON-	PRP
2163	92	14	as	as	IN
2163	92	15	one	one	CD
2163	92	16	of	of	IN
2163	92	17	the	the	DT
2163	92	18	key	key	JJ
2163	92	19	next	next	JJ
2163	92	20	steps	step	NNS
2163	92	21	to	to	TO
2163	92	22	be	be	VB
2163	92	23	tackled	tackle	VBN
2163	92	24	.	.	.
2163	93	1	Consequently	consequently	RB
2163	93	2	,	,	,
2163	93	3	we	-PRON-	PRP
2163	93	4	mention	mention	VBP
2163	93	5	the	the	DT
2163	93	6	research	research	NN
2163	93	7	performed	perform	VBN
2163	93	8	by	by	IN
2163	93	9	Schwartz	Schwartz	NNP
2163	93	10	et	et	FW
2163	93	11	al	al	NNP
2163	93	12	.	.	.
2163	94	1	27	27	CD
2163	94	2	Teufel	Teufel	NNP
2163	94	3	et	et	NNP
2163	94	4	al	al	NNP
2163	94	5	.	.	NNP
2163	94	6	,	,	,
2163	94	7	28	28	CD
2163	94	8	or	or	CC
2163	94	9	Wu	Wu	NNP
2163	94	10	et	et	NNP
2163	94	11	al	al	NNP
2163	94	12	.	.	.
2163	95	1	29	29	CD
2163	95	2	that	that	DT
2163	95	3	deal	deal	VBP
2163	95	4	with	with	IN
2163	95	5	using	use	VBG
2163	95	6	citation	citation	NN
2163	95	7	contexts	contexts	NN
2163	95	8	for	for	IN
2163	95	9	discerning	discern	VBG
2163	95	10	a	a	DT
2163	95	11	citation	citation	NN
2163	95	12	's	's	POS
2163	95	13	function	function	NN
2163	95	14	and	and	CC
2163	95	15	analyzing	analyze	VBG
2163	95	16	how	how	WRB
2163	95	17	this	this	DT
2163	95	18	influences	influence	NNS
2163	95	19	or	or	CC
2163	95	20	is	be	VBZ
2163	95	21	influenced	influence	VBN
2163	95	22	by	by	IN
2163	95	23	the	the	DT
2163	95	24	work	work	NN
2163	95	25	it	-PRON-	PRP
2163	95	26	points	point	VBZ
2163	95	27	to	to	TO
2163	95	28	.	.	.
2163	96	1	INFORMATION	INFORMATION	NNP
2163	96	2	TECHNOLOGY	technology	NN
2163	96	3	AND	and	CC
2163	96	4	LIBRARIES	LIBRARIES	NNP
2163	96	5	|	|	NNP
2163	96	6	JUNE	JUNE	NNP
2163	96	7	2012	2012	CD
2163	96	8	11	11	CD
2163	96	9	3	3	CD
2163	96	10	.	.	.
2163	97	1	METHOD	METHOD	NNP
2163	97	2	This	this	DT
2163	97	3	section	section	NN
2163	97	4	presents	present	VBZ
2163	97	5	the	the	DT
2163	97	6	CRF	CRF	NNP
2163	97	7	chunker	chunker	NN
2163	97	8	model	model	NN
2163	97	9	.	.	.
2163	98	1	We	-PRON-	PRP
2163	98	2	start	start	VBP
2163	98	3	by	by	IN
2163	98	4	defining	define	VBG
2163	98	5	the	the	DT
2163	98	6	preprocessing	preprocesse	VBG
2163	98	7	steps	step	NNS
2163	98	8	that	that	WDT
2163	98	9	deal	deal	VBP
2163	98	10	with	with	IN
2163	98	11	the	the	DT
2163	98	12	extraction	extraction	NN
2163	98	13	of	of	IN
2163	98	14	the	the	DT
2163	98	15	references	reference	NNS
2163	98	16	block	block	NN
2163	98	17	,	,	,
2163	98	18	dividing	divide	VBG
2163	98	19	the	the	DT
2163	98	20	block	block	NN
2163	98	21	into	into	IN
2163	98	22	actual	actual	JJ
2163	98	23	reference	reference	NN
2163	98	24	entries	entry	NNS
2163	98	25	and	and	CC
2163	98	26	cleaning	clean	VBG
2163	98	27	the	the	DT
2163	98	28	reference	reference	NN
2163	98	29	strings	string	NNS
2163	98	30	,	,	,
2163	98	31	and	and	CC
2163	98	32	then	then	RB
2163	98	33	detail	detail	VB
2163	98	34	the	the	DT
2163	98	35	CRF	CRF	NNP
2163	98	36	reference	reference	NN
2163	98	37	chunker	chunker	NN
2163	98	38	features	feature	NNS
2163	98	39	.	.	.
2163	99	1	3.1	3.1	CD
2163	99	2	Prerequisites	prerequisite	NNS
2163	99	3	Most	Most	JJS
2163	99	4	of	of	IN
2163	99	5	the	the	DT
2163	99	6	features	feature	NNS
2163	99	7	used	use	VBN
2163	99	8	by	by	IN
2163	99	9	the	the	DT
2163	99	10	CRF	CRF	NNP
2163	99	11	chunker	chunker	NN
2163	99	12	require	require	VBP
2163	99	13	some	some	DT
2163	99	14	forms	form	NNS
2163	99	15	of	of	IN
2163	99	16	vocabulary	vocabulary	JJ
2163	99	17	entries	entry	NNS
2163	99	18	.	.	.
2163	100	1	Therefore	therefore	RB
2163	100	2	,	,	,
2163	100	3	we	-PRON-	PRP
2163	100	4	have	have	VBP
2163	100	5	manually	manually	RB
2163	100	6	compiled	compile	VBN
2163	100	7	a	a	DT
2163	100	8	comprehensive	comprehensive	JJ
2163	100	9	list	list	NN
2163	100	10	of	of	IN
2163	100	11	gazetteers	gazetteer	NNS
2163	100	12	(	(	-LRB-
2163	100	13	only	only	RB
2163	100	14	for	for	IN
2163	100	15	English	English	NNP
2163	100	16	,	,	,
2163	100	17	except	except	IN
2163	100	18	for	for	IN
2163	100	19	the	the	DT
2163	100	20	names	name	NNS
2163	100	21	)	)	-RRB-
2163	100	22	,	,	,
2163	100	23	explained	explain	VBD
2163	100	24	as	as	IN
2163	100	25	follows	follow	VBZ
2163	100	26	:	:	:
2163	100	27			CD
2163	100	28	FirstName—25,155	FirstName—25,155	NNP
2163	100	29	entries	entry	NNS
2163	100	30	gazetteer	gazetteer	NN
2163	100	31	of	of	IN
2163	100	32	the	the	DT
2163	100	33	most	most	RBS
2163	100	34	common	common	JJ
2163	100	35	first	first	JJ
2163	100	36	names	name	NNS
2163	100	37	(	(	-LRB-
2163	100	38	independent	independent	JJ
2163	100	39	of	of	IN
2163	100	40	gender	gender	NN
2163	100	41	)	)	-RRB-
2163	100	42	;	;	:
2163	100	43			:
2163	100	44	LastName—48,378	LastName—48,378	NNP
2163	100	45	entries	entries	NNP
2163	100	46	list	list	NN
2163	100	47	of	of	IN
2163	100	48	the	the	DT
2163	100	49	most	most	RBS
2163	100	50	common	common	JJ
2163	100	51	surnames	surname	NNS
2163	100	52	;	;	:
2163	100	53			:
2163	100	54	Month	month	NN
2163	100	55	—	—	:
2163	100	56	month	month	NN
2163	100	57	names	name	NNS
2163	100	58	gazetteer	gazetteer	NN
2163	100	59	and	and	CC
2163	100	60	associated	associated	NNP
2163	100	61	abbreviations	abbreviation	NNS
2163	100	62	;	;	,
2163	100	63			:
2163	100	64	VenueType	VenueType	NNP
2163	100	65	—	—	:
2163	100	66	a	a	DT
2163	100	67	structured	structured	JJ
2163	100	68	gazetteer	gazetteer	NN
2163	100	69	with	with	IN
2163	100	70	five	five	CD
2163	100	71	categories	category	NNS
2163	100	72	:	:	:
2163	100	73	Conference	Conference	NNP
2163	100	74	,	,	,
2163	100	75	Workshop	Workshop	NNP
2163	100	76	,	,	,
2163	100	77	Journal	Journal	NNP
2163	100	78	,	,	,
2163	100	79	TechReport	TechReport	NNP
2163	100	80	,	,	,
2163	100	81	and	and	CC
2163	100	82	Website	website	NN
2163	100	83	.	.	.
2163	101	1	Each	each	DT
2163	101	2	category	category	NN
2163	101	3	has	have	VBZ
2163	101	4	attached	attach	VBN
2163	101	5	its	-PRON-	PRP$
2163	101	6	own	own	JJ
2163	101	7	gazetteer	gazetteer	NN
2163	101	8	,	,	,
2163	101	9	containing	contain	VBG
2163	101	10	specific	specific	JJ
2163	101	11	keywords	keyword	NNS
2163	101	12	and	and	CC
2163	101	13	not	not	RB
2163	101	14	actual	actual	JJ
2163	101	15	titles	title	NNS
2163	101	16	.	.	.
2163	102	1	For	for	IN
2163	102	2	example	example	NN
2163	102	3	,	,	,
2163	102	4	the	the	DT
2163	102	5	Conference	Conference	NNP
2163	102	6	gazetteerfeatures	gazetteerfeature	VBZ
2163	102	7	ten	ten	CD
2163	102	8	unigrams	unigram	NNS
2163	102	9	signaling	signal	VBG
2163	102	10	conferences	conference	NNS
2163	102	11	,	,	,
2163	102	12	such	such	JJ
2163	102	13	as	as	IN
2163	102	14	Conference	Conference	NNP
2163	102	15	,	,	,
2163	102	16	Conf	Conf	NNP
2163	102	17	,	,	,
2163	102	18	or	or	CC
2163	102	19	Symposium	Symposium	NNP
2163	102	20	;	;	:
2163	102	21			NFP
2163	102	22	Location	location	NN
2163	102	23	—	—	:
2163	102	24	places	place	NNS
2163	102	25	,	,	,
2163	102	26	cities	city	NNS
2163	102	27	,	,	,
2163	102	28	and	and	CC
2163	102	29	countries	country	NNS
2163	102	30	gazetteer	gazetteer	NN
2163	102	31	comprising	comprise	VBG
2163	102	32	17,336	17,336	CD
2163	102	33	entries	entry	NNS
2163	102	34	;	;	:
2163	102	35			NFP
2163	102	36	Organization—150	organization—150	ADD
2163	102	37	entries	entries	NNP
2163	102	38	gazetteer	gazetteer	NNP
2163	102	39	listing	list	VBG
2163	102	40	organization	organization	NN
2163	102	41	prefixes	prefix	NNS
2163	102	42	and	and	CC
2163	102	43	suffixes	suffix	NNS
2163	102	44	(	(	-LRB-
2163	102	45	e.g.	e.g.	RB
2163	102	46	,	,	,
2163	102	47	e	e	NNP
2163	102	48	.	.	.
2163	102	49	V.	V.	NNP
2163	103	1	or	or	CC
2163	103	2	KGaA	KGaA	NNP
2163	103	3	)	)	-RRB-
2163	103	4	;	;	:
2163	103	5			NFP
2163	103	6	Proceedings	proceeding	NNS
2163	103	7	—	—	:
2163	103	8	simple	simple	JJ
2163	103	9	list	list	NN
2163	103	10	of	of	IN
2163	103	11	all	all	DT
2163	103	12	possible	possible	JJ
2163	103	13	appearances	appearance	NNS
2163	103	14	of	of	IN
2163	103	15	the	the	DT
2163	103	16	Proceedings	Proceedings	NNP
2163	103	17	marker	marker	NN
2163	103	18	;	;	:
2163	103	19			NN
2163	103	20	Publisher—564	Publisher—564	NNP
2163	103	21	entries	entries	NNP
2163	103	22	gazetteer	gazetteer	NNP
2163	103	23	comprising	comprise	VBG
2163	103	24	publisher	publisher	NN
2163	103	25	unigrams	unigrams	NNP
2163	103	26	(	(	-LRB-
2163	103	27	produced	produce	VBN
2163	103	28	from	from	IN
2163	103	29	around	around	IN
2163	103	30	150	150	CD
2163	103	31	publisher	publisher	NN
2163	103	32	names	name	NNS
2163	103	33	)	)	-RRB-
2163	103	34	;	;	:
2163	103	35			:
2163	103	36	JTitle—12,101	jtitle—12,101	CD
2163	103	37	entries	entry	NNS
2163	103	38	list	list	NN
2163	103	39	of	of	IN
2163	103	40	journal	journal	NNP
2163	103	41	title	title	NNP
2163	103	42	unigrams	unigrams	NNP
2163	103	43	(	(	-LRB-
2163	103	44	produced	produce	VBN
2163	103	45	from	from	IN
2163	103	46	around	around	IN
2163	103	47	1600	1600	CD
2163	103	48	journal	journal	NNP
2163	103	49	titles	title	NNS
2163	103	50	)	)	-RRB-
2163	103	51	;	;	:
2163	103	52			:
2163	103	53	Connection	connection	NN
2163	103	54	—	—	:
2163	103	55	a	a	DT
2163	103	56	42	42	CD
2163	103	57	entries	entry	NNS
2163	103	58	stop	stop	NN
2163	103	59	-	-	HYPH
2163	103	60	word	word	NN
2163	103	61	gazetteer	gazetteer	NN
2163	103	62	(	(	-LRB-
2163	103	63	e.g.	e.g.	RB
2163	103	64	,	,	,
2163	103	65	to	to	IN
2163	103	66	,	,	,
2163	103	67	and	and	CC
2163	103	68	,	,	,
2163	103	69	as	as	IN
2163	103	70	)	)	-RRB-
2163	103	71	.	.	.
2163	104	1	3.2	3.2	CD
2163	104	2	Preprocessing	Preprocessing	NNP
2163	104	3	In	in	IN
2163	104	4	the	the	DT
2163	104	5	preprocessing	preprocessing	NN
2163	104	6	stage	stage	NN
2163	104	7	we	-PRON-	PRP
2163	104	8	deal	deal	VBP
2163	104	9	with	with	IN
2163	104	10	three	three	CD
2163	104	11	aspects	aspect	NNS
2163	104	12	:	:	:
2163	104	13			:
2163	104	14	cleaning	clean	VBG
2163	104	15	the	the	DT
2163	104	16	provided	provide	VBN
2163	104	17	input	input	NN
2163	104	18	,	,	,
2163	104	19			:
2163	104	20	extracting	extract	VBG
2163	104	21	the	the	DT
2163	104	22	reference	reference	NN
2163	104	23	block	block	NN
2163	104	24	,	,	,
2163	104	25	and	and	CC
2163	104	26			NFP
2163	104	27	the	the	DT
2163	104	28	division	division	NN
2163	104	29	of	of	IN
2163	104	30	the	the	DT
2163	104	31	reference	reference	NN
2163	104	32	block	block	NN
2163	104	33	into	into	IN
2163	104	34	reference	reference	NN
2163	104	35	entries	entry	NNS
2163	104	36	.	.	.
2163	105	1	The	the	DT
2163	105	2	first	first	JJ
2163	105	3	step	step	NN
2163	105	4	aims	aim	VBZ
2163	105	5	to	to	TO
2163	105	6	clean	clean	VB
2163	105	7	the	the	DT
2163	105	8	raw	raw	JJ
2163	105	9	textual	textual	JJ
2163	105	10	input	input	NN
2163	105	11	received	receive	VBN
2163	105	12	by	by	IN
2163	105	13	the	the	DT
2163	105	14	chunker	chunker	NN
2163	105	15	of	of	IN
2163	105	16	unwanted	unwanted	JJ
2163	105	17	spacing	space	VBG
2163	105	18	characters	character	NNS
2163	105	19	while	while	IN
2163	105	20	at	at	IN
2163	105	21	the	the	DT
2163	105	22	same	same	JJ
2163	105	23	time	time	NN
2163	105	24	ensuring	ensure	VBG
2163	105	25	proper	proper	JJ
2163	105	26	spacing	spacing	NN
2163	105	27	where	where	WRB
2163	105	28	necessary	necessary	JJ
2163	105	29	.	.	.
2163	106	1	Since	since	IN
2163	106	2	the	the	DT
2163	106	3	source	source	NN
2163	106	4	of	of	IN
2163	106	5	the	the	DT
2163	106	6	textual	textual	JJ
2163	106	7	input	input	NN
2163	106	8	is	be	VBZ
2163	106	9	unknown	unknown	JJ
2163	106	10	to	to	IN
2163	106	11	the	the	DT
2163	106	12	chunker	chunker	NN
2163	106	13	,	,	,
2163	106	14	we	-PRON-	PRP
2163	106	15	make	make	VBP
2163	106	16	no	no	DT
2163	106	17	assumptions	assumption	NNS
2163	106	18	with	with	IN
2163	106	19	regard	regard	NN
2163	106	20	to	to	IN
2163	106	21	its	-PRON-	PRP$
2163	106	22	structure	structure	NN
2163	106	23	or	or	CC
2163	106	24	content	content	NN
2163	106	25	.	.	.
2163	107	1	30	30	CD
2163	107	2	Thus	thus	RB
2163	107	3	,	,	,
2163	107	4	in	in	IN
2163	107	5	order	order	NN
2163	107	6	to	to	TO
2163	107	7	avoid	avoid	VB
2163	107	8	inherent	inherent	JJ
2163	107	9	errors	error	NNS
2163	107	10	that	that	WDT
2163	107	11	might	may	MD
2163	107	12	appear	appear	VB
2163	107	13	as	as	IN
2163	107	14	a	a	DT
2163	107	15	result	result	NN
2163	107	16	of	of	IN
2163	107	17	extracting	extract	VBG
2163	107	18	the	the	DT
2163	107	19	raw	raw	JJ
2163	107	20	text	text	NN
2163	107	21	from	from	IN
2163	107	22	the	the	DT
2163	107	23	original	original	JJ
2163	107	24	document	document	NN
2163	107	25	,	,	,
2163	107	26	we	-PRON-	PRP
2163	107	27	perform	perform	VBP
2163	107	28	the	the	DT
2163	107	29	following	follow	VBG
2163	107	30	cleaning	cleaning	NN
2163	107	31	steps	step	NNS
2163	107	32	:	:	:
2163	107	33			NFP
2163	107	34	we	-PRON-	PRP
2163	107	35	compress	compress	VBP
2163	107	36	the	the	DT
2163	107	37	text	text	NN
2163	107	38	by	by	IN
2163	107	39	eliminating	eliminate	VBG
2163	107	40	unnecessary	unnecessary	JJ
2163	107	41	carriage	carriage	NN
2163	107	42	returns	return	NNS
2163	107	43	,	,	,
2163	107	44	such	such	JJ
2163	107	45	that	that	IN
2163	107	46	the	the	DT
2163	107	47	lines	line	NNS
2163	107	48	containing	contain	VBG
2163	107	49	less	less	JJR
2163	107	50	than	than	IN
2163	107	51	15	15	CD
2163	107	52	characters	character	NNS
2163	107	53	are	be	VBP
2163	107	54	merged	merge	VBN
2163	107	55	with	with	IN
2163	107	56	previous	previous	JJ
2163	107	57	ones	one	NNS
2163	107	58	,	,	,
2163	107	59	31	31	CD
2163	107	60			:
2163	107	61	we	-PRON-	PRP
2163	107	62	introduce	introduce	VBP
2163	107	63	spaces	space	NNS
2163	107	64	after	after	IN
2163	107	65	some	some	DT
2163	107	66	punctuation	punctuation	NN
2163	107	67	characters	character	NNS
2163	107	68	,	,	,
2163	107	69	such	such	JJ
2163	107	70	as	as	IN
2163	107	71	“	"	``
2163	107	72	,	,	,
2163	107	73	,	,	,
2163	107	74	”	"	''
2163	107	75	“	"	``
2163	107	76	.	.	.
2163	107	77	”	"	''
2163	107	78	or	or	CC
2163	107	79	“	"	``
2163	107	80	-	-	:
2163	107	81	”	"	''
2163	107	82	,	,	,
2163	107	83	and	and	CC
2163	107	84	finally	finally	RB
2163	107	85	,	,	,
2163	107	86			NFP
2163	107	87	we	-PRON-	PRP
2163	107	88	split	split	VBD
2163	107	89	the	the	DT
2163	107	90	camel	camel	NN
2163	107	91	-	-	HYPH
2163	107	92	cased	case	VBN
2163	107	93	strings	string	NNS
2163	107	94	,	,	,
2163	107	95	such	such	JJ
2163	107	96	as	as	IN
2163	107	97	JohnDoe	JohnDoe	NNP
2163	107	98	.	.	.
2163	108	1	REFERENCE	reference	NN
2163	108	2	INFORMATION	INFORMATION	NNS
2163	108	3	EXTRACTION	extraction	NN
2163	108	4	AND	and	CC
2163	108	5	PROCESSING	processing	NN
2163	108	6	|GROZA	|groza	NN
2163	108	7	,	,	,
2163	108	8	GRIMNES	GRIMNES	NNP
2163	108	9	,	,	,
2163	108	10	AND	and	CC
2163	108	11	HANDSCHUH	handschuh	NN
2163	108	12	12	12	CD
2163	108	13	The	the	DT
2163	108	14	result	result	NN
2163	108	15	will	will	MD
2163	108	16	be	be	VB
2163	108	17	a	a	DT
2163	108	18	compact	compact	JJ
2163	108	19	and	and	CC
2163	108	20	clean	clean	JJ
2163	108	21	version	version	NN
2163	108	22	of	of	IN
2163	108	23	the	the	DT
2163	108	24	input	input	NN
2163	108	25	.	.	.
2163	109	1	Also	also	RB
2163	109	2	,	,	,
2163	109	3	if	if	IN
2163	109	4	the	the	DT
2163	109	5	raw	raw	JJ
2163	109	6	input	input	NN
2163	109	7	is	be	VBZ
2163	109	8	already	already	RB
2163	109	9	compact	compact	JJ
2163	109	10	and	and	CC
2163	109	11	clean	clean	JJ
2163	109	12	,	,	,
2163	109	13	this	this	DT
2163	109	14	preprocessing	preprocesse	VBG
2163	109	15	step	step	NN
2163	109	16	will	will	MD
2163	109	17	not	not	RB
2163	109	18	affect	affect	VB
2163	109	19	it	-PRON-	PRP
2163	109	20	.	.	.
2163	110	1	The	the	DT
2163	110	2	extraction	extraction	NN
2163	110	3	of	of	IN
2163	110	4	the	the	DT
2163	110	5	reference	reference	NN
2163	110	6	block	block	NN
2163	110	7	is	be	VBZ
2163	110	8	done	do	VBN
2163	110	9	using	use	VBG
2163	110	10	regular	regular	JJ
2163	110	11	expressions	expression	NNS
2163	110	12	.	.	.
2163	111	1	Generally	generally	RB
2163	111	2	,	,	,
2163	111	3	we	-PRON-	PRP
2163	111	4	search	search	VBP
2163	111	5	in	in	IN
2163	111	6	the	the	DT
2163	111	7	compacted	compact	VBN
2163	111	8	and	and	CC
2163	111	9	cleaned	clean	VBN
2163	111	10	input	input	NN
2163	111	11	for	for	IN
2163	111	12	specific	specific	JJ
2163	111	13	markers	marker	NNS
2163	111	14	,	,	,
2163	111	15	like	like	IN
2163	111	16	References	References	NNPS
2163	111	17	or	or	CC
2163	111	18	Bibliography	Bibliography	NNP
2163	111	19	,	,	,
2163	111	20	located	locate	VBN
2163	111	21	mainly	mainly	RB
2163	111	22	at	at	IN
2163	111	23	the	the	DT
2163	111	24	beginning	beginning	NN
2163	111	25	of	of	IN
2163	111	26	a	a	DT
2163	111	27	line	line	NN
2163	111	28	.	.	.
2163	112	1	If	if	IN
2163	112	2	these	these	DT
2163	112	3	are	be	VBP
2163	112	4	not	not	RB
2163	112	5	directly	directly	RB
2163	112	6	found	find	VBN
2163	112	7	,	,	,
2163	112	8	we	-PRON-	PRP
2163	112	9	try	try	VBP
2163	112	10	different	different	JJ
2163	112	11	variations	variation	NNS
2163	112	12	,	,	,
2163	112	13	such	such	JJ
2163	112	14	as	as	IN
2163	112	15	,	,	,
2163	112	16	looking	look	VBG
2163	112	17	for	for	IN
2163	112	18	the	the	DT
2163	112	19	markers	marker	NNS
2163	112	20	at	at	IN
2163	112	21	the	the	DT
2163	112	22	end	end	NN
2163	112	23	of	of	IN
2163	112	24	a	a	DT
2163	112	25	line	line	NN
2163	112	26	,	,	,
2163	112	27	or	or	CC
2163	112	28	looking	look	VBG
2163	112	29	for	for	IN
2163	112	30	split	split	NN
2163	112	31	markers	marker	NNS
2163	112	32	onto	onto	IN
2163	112	33	two	two	CD
2163	112	34	lines	line	NNS
2163	112	35	(	(	-LRB-
2163	112	36	e.g.	e.g.	RB
2163	112	37	,	,	,
2163	112	38	Ref	Ref	NNP
2163	112	39	–	–	:
2163	112	40	erences	erence	NNS
2163	112	41	,	,	,
2163	112	42	or	or	CC
2163	112	43	Refer	refer	VB
2163	112	44	–	–	:
2163	112	45	ences	ence	NNS
2163	112	46	)	)	-RRB-
2163	112	47	.	.	.
2163	113	1	This	this	DT
2163	113	2	latter	latter	JJ
2163	113	3	case	case	NN
2163	113	4	is	be	VBZ
2163	113	5	a	a	DT
2163	113	6	typical	typical	JJ
2163	113	7	consequence	consequence	NN
2163	113	8	of	of	IN
2163	113	9	the	the	DT
2163	113	10	above	above	RB
2163	113	11	-	-	HYPH
2163	113	12	described	describe	VBN
2163	113	13	compacting	compact	VBG
2163	113	14	step	step	NN
2163	113	15	if	if	IN
2163	113	16	the	the	DT
2163	113	17	initial	initial	JJ
2163	113	18	input	input	NN
2163	113	19	was	be	VBD
2163	113	20	erroneously	erroneously	RB
2163	113	21	extracted	extract	VBN
2163	113	22	.	.	.
2163	114	1	The	the	DT
2163	114	2	text	text	NN
2163	114	3	following	follow	VBG
2163	114	4	the	the	DT
2163	114	5	markers	marker	NNS
2163	114	6	is	be	VBZ
2163	114	7	considered	consider	VBN
2163	114	8	for	for	IN
2163	114	9	division	division	NN
2163	114	10	,	,	,
2163	114	11	although	although	IN
2163	114	12	it	-PRON-	PRP
2163	114	13	may	may	MD
2163	114	14	contain	contain	VB
2163	114	15	unwanted	unwanted	JJ
2163	114	16	parts	part	NNS
2163	114	17	such	such	JJ
2163	114	18	as	as	IN
2163	114	19	appendices	appendix	NNS
2163	114	20	or	or	CC
2163	114	21	tables	table	NNS
2163	114	22	.	.	.
2163	115	1	The	the	DT
2163	115	2	division	division	NN
2163	115	3	into	into	IN
2163	115	4	individual	individual	JJ
2163	115	5	reference	reference	NN
2163	115	6	entries	entry	NNS
2163	115	7	is	be	VBZ
2163	115	8	performed	perform	VBN
2163	115	9	on	on	IN
2163	115	10	a	a	DT
2163	115	11	case	case	NN
2163	115	12	basis	basis	NN
2163	115	13	.	.	.
2163	116	1	After	after	IN
2163	116	2	splitting	split	VBG
2163	116	3	the	the	DT
2163	116	4	reference	reference	NN
2163	116	5	block	block	NN
2163	116	6	based	base	VBN
2163	116	7	on	on	IN
2163	116	8	new	new	JJ
2163	116	9	lines	line	NNS
2163	116	10	,	,	,
2163	116	11	we	-PRON-	PRP
2163	116	12	look	look	VBP
2163	116	13	for	for	IN
2163	116	14	prefix	prefix	JJ
2163	116	15	patterns	pattern	NNS
2163	116	16	at	at	IN
2163	116	17	the	the	DT
2163	116	18	beginning	beginning	NN
2163	116	19	of	of	IN
2163	116	20	each	each	DT
2163	116	21	line	line	NN
2163	116	22	.	.	.
2163	117	1	As	as	IN
2163	117	2	an	an	DT
2163	117	3	example	example	NN
2163	117	4	,	,	,
2163	117	5	we	-PRON-	PRP
2163	117	6	analyze	analyze	VBP
2163	117	7	which	which	WDT
2163	117	8	lines	line	NNS
2163	117	9	start	start	VBP
2163	117	10	with	with	IN
2163	117	11	“	"	``
2163	117	12	[	[	-LRB-
2163	117	13	”	"	''
2163	117	14	,	,	,
2163	117	15	“	"	``
2163	117	16	(	(	-LRB-
2163	117	17	”	"	''
2163	117	18	,	,	,
2163	117	19	or	or	CC
2163	117	20	a	a	DT
2163	117	21	number	number	NN
2163	117	22	followed	follow	VBN
2163	117	23	by	by	IN
2163	117	24	“	"	``
2163	117	25	.	.	.
2163	117	26	”	"	''
2163	117	27	or	or	CC
2163	117	28	space	space	NN
2163	117	29	,	,	,
2163	117	30	and	and	CC
2163	117	31	we	-PRON-	PRP
2163	117	32	record	record	VBP
2163	117	33	the	the	DT
2163	117	34	positions	position	NNS
2163	117	35	of	of	IN
2163	117	36	these	these	DT
2163	117	37	lines	line	NNS
2163	117	38	in	in	IN
2163	117	39	the	the	DT
2163	117	40	list	list	NN
2163	117	41	of	of	IN
2163	117	42	all	all	DT
2163	117	43	lines	line	NNS
2163	117	44	.	.	.
2163	118	1	To	to	TO
2163	118	2	ensure	ensure	VB
2163	118	3	that	that	IN
2163	118	4	we	-PRON-	PRP
2163	118	5	do	do	VBP
2163	118	6	n't	not	RB
2163	118	7	consider	consider	VB
2163	118	8	any	any	DT
2163	118	9	false	false	JJ
2163	118	10	positives	positive	NNS
2163	118	11	when	when	WRB
2163	118	12	merging	merge	VBG
2163	118	13	the	the	DT
2163	118	14	adjacent	adjacent	JJ
2163	118	15	lines	line	NNS
2163	118	16	into	into	IN
2163	118	17	a	a	DT
2163	118	18	reference	reference	NN
2163	118	19	entry	entry	NN
2163	118	20	,	,	,
2163	118	21	we	-PRON-	PRP
2163	118	22	compute	compute	VBP
2163	118	23	a	a	DT
2163	118	24	global	global	JJ
2163	118	25	average	average	NN
2163	118	26	of	of	IN
2163	118	27	the	the	DT
2163	118	28	differences	difference	NNS
2163	118	29	between	between	IN
2163	118	30	positions	position	NNS
2163	118	31	.	.	.
2163	119	1	Assuming	assume	VBG
2163	119	2	that	that	IN
2163	119	3	a	a	DT
2163	119	4	reference	reference	NN
2163	119	5	does	do	VBZ
2163	119	6	not	not	RB
2163	119	7	span	span	VB
2163	119	8	on	on	IN
2163	119	9	more	more	JJR
2163	119	10	than	than	IN
2163	119	11	four	four	CD
2163	119	12	lines	line	NNS
2163	119	13	,	,	,
2163	119	14	if	if	IN
2163	119	15	this	this	DT
2163	119	16	average	average	NN
2163	119	17	is	be	VBZ
2163	119	18	between	between	IN
2163	119	19	one	one	CD
2163	119	20	and	and	CC
2163	119	21	four	four	CD
2163	119	22	,	,	,
2163	119	23	a	a	DT
2163	119	24	reference	reference	NN
2163	119	25	entry	entry	NN
2163	119	26	is	be	VBZ
2163	119	27	created	create	VBN
2163	119	28	.	.	.
2163	120	1	The	the	DT
2163	120	2	same	same	JJ
2163	120	3	average	average	NN
2163	120	4	is	be	VBZ
2163	120	5	also	also	RB
2163	120	6	used	use	VBN
2163	120	7	to	to	TO
2163	120	8	extract	extract	VB
2163	120	9	the	the	DT
2163	120	10	last	last	JJ
2163	120	11	reference	reference	NN
2163	120	12	in	in	IN
2163	120	13	the	the	DT
2163	120	14	list	list	NN
2163	120	15	,	,	,
2163	120	16	thus	thus	RB
2163	120	17	detaching	detach	VBG
2163	120	18	it	-PRON-	PRP
2163	120	19	from	from	IN
2163	120	20	eventual	eventual	JJ
2163	120	21	appendices	appendix	NNS
2163	120	22	or	or	CC
2163	120	23	tables	table	NNS
2163	120	24	.	.	.
2163	121	1	3.3	3.3	CD
2163	121	2	The	the	DT
2163	121	3	reference	reference	NN
2163	121	4	chunking	chunk	VBG
2163	121	5	model	model	NN
2163	121	6	We	-PRON-	PRP
2163	121	7	have	have	VBP
2163	121	8	built	build	VBN
2163	121	9	the	the	DT
2163	121	10	CRF	CRF	NNP
2163	121	11	learning	learn	VBG
2163	121	12	model	model	NN
2163	121	13	based	base	VBN
2163	121	14	on	on	IN
2163	121	15	a	a	DT
2163	121	16	series	series	NN
2163	121	17	of	of	IN
2163	121	18	features	feature	NNS
2163	121	19	used	use	VBN
2163	121	20	in	in	IN
2163	121	21	principle	principle	NN
2163	121	22	also	also	RB
2163	121	23	by	by	IN
2163	121	24	the	the	DT
2163	121	25	other	other	JJ
2163	121	26	CRF	CRF	NNP
2163	121	27	reference	reference	NN
2163	121	28	chunking	chunk	VBG
2163	121	29	approaches	approach	NNS
2163	121	30	such	such	JJ
2163	121	31	as	as	IN
2163	121	32	ParsCit	ParsCit	NNP
2163	121	33	32	32	CD
2163	121	34	or	or	CC
2163	121	35	Peng	Peng	NNP
2163	121	36	and	and	CC
2163	121	37	McCallum	McCallum	NNP
2163	121	38	33	33	CD
2163	121	39	.	.	.
2163	122	1	A	a	DT
2163	122	2	set	set	NN
2163	122	3	of	of	IN
2163	122	4	feature	feature	NN
2163	122	5	values	value	NNS
2163	122	6	is	be	VBZ
2163	122	7	used	use	VBN
2163	122	8	to	to	TO
2163	122	9	characterize	characterize	VB
2163	122	10	each	each	DT
2163	122	11	token	token	JJ
2163	122	12	present	present	JJ
2163	122	13	in	in	IN
2163	122	14	the	the	DT
2163	122	15	reference	reference	NN
2163	122	16	string	string	NN
2163	122	17	,	,	,
2163	122	18	where	where	WRB
2163	122	19	the	the	DT
2163	122	20	reference	reference	NN
2163	122	21	's	's	POS
2163	122	22	token	token	JJ
2163	122	23	list	list	NN
2163	122	24	is	be	VBZ
2163	122	25	obtained	obtain	VBN
2163	122	26	by	by	IN
2163	122	27	dividing	divide	VBG
2163	122	28	the	the	DT
2163	122	29	reference	reference	NN
2163	122	30	string	string	NN
2163	122	31	into	into	IN
2163	122	32	space	space	NN
2163	122	33	-	-	HYPH
2163	122	34	separated	separate	VBN
2163	122	35	pieces	piece	NNS
2163	122	36	.	.	.
2163	123	1	The	the	DT
2163	123	2	complete	complete	JJ
2163	123	3	list	list	NN
2163	123	4	of	of	IN
2163	123	5	features	feature	NNS
2163	123	6	is	be	VBZ
2163	123	7	detailed	detail	VBN
2163	123	8	as	as	IN
2163	123	9	follows	follow	VBZ
2163	123	10	.	.	.
2163	124	1	We	-PRON-	PRP
2163	124	2	use	use	VBP
2163	124	3	example	example	NN
2163	124	4	1	1	CD
2163	124	5	from	from	IN
2163	124	6	figure	figure	NN
2163	124	7	1	1	CD
2163	124	8	toexemplify	toexemplify	VB
2163	124	9	the	the	DT
2163	124	10	feature	feature	NN
2163	124	11	values	value	NNS
2163	124	12	.	.	.
2163	125	1			:
2163	125	2	Token	Token	NNP
2163	125	3	—	—	:
2163	125	4	the	the	DT
2163	125	5	original	original	JJ
2163	125	6	reference	reference	NN
2163	125	7	token	token	NN
2163	125	8	:	:	:
2163	125	9	Bronzwaer	Bronzwaer	NNP
2163	125	10	,	,	,
2163	125	11			:
2163	125	12	Clean	clean	JJ
2163	125	13	token	token	NN
2163	125	14	—	—	:
2163	125	15	the	the	DT
2163	125	16	original	original	JJ
2163	125	17	token	token	NN
2163	125	18	,	,	,
2163	125	19	stripped	strip	VBN
2163	125	20	of	of	IN
2163	125	21	any	any	DT
2163	125	22	punctuation	punctuation	NN
2163	125	23	and	and	CC
2163	125	24	lower	low	JJR
2163	125	25	cased	case	VBN
2163	125	26	:	:	:
2163	125	27	bronzwaer	bronzwaer	NNP
2163	125	28			:
2163	125	29	Token	token	JJ
2163	125	30	ending	ending	NN
2163	125	31	—	—	:
2163	125	32	a	a	DT
2163	125	33	flag	flag	NN
2163	125	34	signaling	signal	VBG
2163	125	35	the	the	DT
2163	125	36	type	type	NN
2163	125	37	of	of	IN
2163	125	38	ending	end	VBG
2163	125	39	(	(	-LRB-
2163	125	40	possible	possible	JJ
2163	125	41	values	value	NNS
2163	125	42	:	:	:
2163	125	43	lower	low	JJR
2163	125	44	cap	cap	NN
2163	125	45	–	–	:
2163	125	46	c	c	NN
2163	125	47	/	/	SYM
2163	125	48	upper	upper	JJ
2163	125	49	cap	cap	NN
2163	125	50	–	–	:
2163	125	51	C	c	NN
2163	125	52	/	/	SYM
2163	125	53	digit	digit	NN
2163	125	54	–	–	:
2163	125	55	0	0	CD
2163	125	56	/	/	SYM
2163	125	57	punctuation	punctuation	NN
2163	125	58	character	character	NN
2163	125	59	:	:	:
2163	125	60	,	,	,
2163	125	61			:
2163	125	62	Token	token	JJ
2163	125	63	decomposition	decomposition	NN
2163	125	64	–	–	:
2163	125	65	start	start	VB
2163	125	66	—	—	:
2163	125	67	five	five	CD
2163	125	68	individual	individual	JJ
2163	125	69	values	value	NNS
2163	125	70	corresponding	correspond	VBG
2163	125	71	to	to	IN
2163	125	72	token	token	VB
2163	125	73	's	's	POS
2163	125	74	first	first	JJ
2163	125	75	five	five	CD
2163	125	76	characters	character	NNS
2163	125	77	,	,	,
2163	125	78	taken	take	VBN
2163	125	79	gradually	gradually	RB
2163	125	80	:	:	:
2163	125	81	B	b	UH
2163	125	82	,	,	,
2163	125	83	Br	br	UH
2163	125	84	,	,	,
2163	125	85	Bro	Bro	NNP
2163	125	86	,	,	,
2163	125	87	Bron	Bron	NNP
2163	125	88	,	,	,
2163	125	89	Bronz	Bronz	NNP
2163	125	90			JJ
2163	125	91	Token	token	JJ
2163	125	92	decomposition	decomposition	NN
2163	125	93	–	–	:
2163	125	94	end	end	NN
2163	125	95	—	—	:
2163	125	96	five	five	CD
2163	125	97	individual	individual	JJ
2163	125	98	values	value	NNS
2163	125	99	corresponding	correspond	VBG
2163	125	100	to	to	IN
2163	125	101	the	the	DT
2163	125	102	token	token	NN
2163	125	103	's	's	POS
2163	125	104	last	last	JJ
2163	125	105	five	five	CD
2163	125	106	characters	character	NNS
2163	125	107	,	,	,
2163	125	108	taken	take	VBN
2163	125	109	gradually	gradually	RB
2163	125	110	:	:	:
2163	125	111	r	r	LS
2163	125	112	,	,	,
2163	125	113	er	er	UH
2163	125	114	,	,	,
2163	125	115	aer	aer	NNP
2163	125	116	,	,	,
2163	125	117	waer	waer	NNP
2163	125	118	,	,	,
2163	125	119	zwaer	zwaer	NNP
2163	125	120	,	,	,
2163	125	121			:
2163	125	122	POS	POS	NNP
2163	125	123	Tag	tag	NN
2163	125	124	—	—	:
2163	125	125	the	the	DT
2163	125	126	token	token	NN
2163	125	127	's	's	POS
2163	125	128	part	part	NN
2163	125	129	of	of	IN
2163	125	130	speech	speech	NN
2163	125	131	tag	tag	NN
2163	125	132	(	(	-LRB-
2163	125	133	possible	possible	JJ
2163	125	134	values	value	NNS
2163	125	135	:	:	:
2163	125	136	proper	proper	JJ
2163	125	137	noun	noun	JJ
2163	125	138	phrase	phrase	NN
2163	125	139	–	–	:
2163	125	140	NNP	NNP	NNP
2163	125	141	,	,	,
2163	125	142			:
2163	125	143	noun	noun	NNP
2163	125	144	phrase	phrase	NN
2163	125	145	–	–	:
2163	125	146	NP	NP	NNP
2163	125	147	,	,	,
2163	125	148	adjective	adjective	NN
2163	125	149	–	–	:
2163	125	150	JJ	jj	JJ
2163	125	151	,	,	,
2163	125	152	cardinal	cardinal	JJ
2163	125	153	number	number	NN
2163	125	154	–	–	:
2163	125	155	CD	cd	NN
2163	125	156	,	,	,
2163	125	157	etc	etc	FW
2163	125	158	)	)	-RRB-
2163	125	159	:	:	:
2163	125	160	NNP	NNP	NNP
2163	125	161			:
2163	125	162	Orthographic	orthographic	JJ
2163	125	163	case	case	NN
2163	125	164	—	—	:
2163	125	165	a	a	DT
2163	125	166	flag	flag	NN
2163	125	167	signaling	signal	VBG
2163	125	168	the	the	DT
2163	125	169	token	token	NN
2163	125	170	's	's	POS
2163	125	171	orthographic	orthographic	JJ
2163	125	172	case	case	NN
2163	125	173	(	(	-LRB-
2163	125	174	possible	possible	JJ
2163	125	175	values	value	NNS
2163	125	176	:	:	:
2163	125	177			NFP
2163	125	178	initialCap	initialCap	NNP
2163	125	179	,	,	,
2163	125	180	singleCap	singleCap	NNP
2163	125	181	,	,	,
2163	125	182	lowercase	lowercase	NN
2163	125	183	,	,	,
2163	125	184	mixedCaps	mixedCaps	NNP
2163	125	185	,	,	,
2163	125	186	allCaps	allCaps	NNP
2163	125	187	)	)	-RRB-
2163	125	188	:	:	:
2163	125	189	singleCap	singleCap	NNP
2163	125	190			:
2163	125	191	Punctuation	punctuation	NN
2163	125	192	type	type	NN
2163	125	193	—	—	:
2163	125	194	a	a	DT
2163	125	195	flag	flag	NN
2163	125	196	signaling	signal	VBG
2163	125	197	the	the	DT
2163	125	198	presence	presence	NN
2163	125	199	and	and	CC
2163	125	200	type	type	NN
2163	125	201	of	of	IN
2163	125	202	a	a	DT
2163	125	203	trailing	trail	VBG
2163	125	204	punctuation	punctuation	NN
2163	125	205	character	character	NN
2163	125	206	(	(	-LRB-
2163	125	207	possible	possible	JJ
2163	125	208	values	value	NNS
2163	125	209	:	:	:
2163	125	210	cont	cont	NNP
2163	125	211	,	,	,
2163	125	212	stop	stop	VB
2163	125	213	,	,	,
2163	125	214	other	other	JJ
2163	125	215	)	)	-RRB-
2163	125	216	:	:	:
2163	125	217	cont	cont	NN
2163	125	218			:
2163	125	219	Number	number	NN
2163	125	220	type	type	NN
2163	125	221	—	—	:
2163	125	222	a	a	DT
2163	125	223	flag	flag	NN
2163	125	224	signaling	signal	VBG
2163	125	225	the	the	DT
2163	125	226	presence	presence	NN
2163	125	227	and	and	CC
2163	125	228	type	type	NN
2163	125	229	of	of	IN
2163	125	230	a	a	DT
2163	125	231	number	number	NN
2163	125	232	in	in	IN
2163	125	233	the	the	DT
2163	125	234	token	token	JJ
2163	125	235	(	(	-LRB-
2163	125	236	possible	possible	JJ
2163	125	237	values	value	NNS
2163	125	238	:	:	:
2163	125	239	year	year	NN
2163	125	240	,	,	,
2163	125	241	ordinal	ordinal	JJ
2163	125	242	,	,	,
2163	125	243	1dig	1dig	CD
2163	125	244	,	,	,
2163	125	245	2dig	2dig	CD
2163	125	246	,	,	,
2163	125	247	3dig	3dig	CD
2163	125	248	,	,	,
2163	125	249	4dig	4dig	CD
2163	125	250	,	,	,
2163	125	251	4dig+	4dig+	CD
2163	125	252	,	,	,
2163	125	253	noNumber	noNumber	NNP
2163	125	254	)	)	-RRB-
2163	125	255	:	:	:
2163	125	256	noNumber	noNumber	NNP
2163	125	257	INFORMATION	INFORMATION	VBD
2163	125	258	TECHNOLOGY	TECHNOLOGY	NNP
2163	125	259	AND	and	CC
2163	125	260	LIBRARIES	LIBRARIES	NNP
2163	125	261	|	|	NNP
2163	125	262	JUNE	JUNE	NNP
2163	125	263	2012	2012	CD
2163	125	264	13	13	CD
2163	125	265			:
2163	125	266	Dictionary	Dictionary	NNP
2163	125	267	entries	entry	NNS
2163	125	268	—	—	:
2163	125	269	a	a	DT
2163	125	270	set	set	NN
2163	125	271	of	of	IN
2163	125	272	ten	ten	CD
2163	125	273	flags	flag	NNS
2163	125	274	signaling	signal	VBG
2163	125	275	the	the	DT
2163	125	276	presence	presence	NN
2163	125	277	of	of	IN
2163	125	278	the	the	DT
2163	125	279	token	token	NN
2163	125	280	in	in	IN
2163	125	281	the	the	DT
2163	125	282	set	set	NN
2163	125	283	of	of	IN
2163	125	284	individual	individual	JJ
2163	125	285	gazetteers	gazetteer	NNS
2163	125	286	listed	list	VBN
2163	125	287	in	in	IN
2163	125	288	Sect	Sect	NNP
2163	125	289	.	.	.
2163	126	1	3.1	3.1	CD
2163	126	2	.	.	.
2163	127	1	For	for	IN
2163	127	2	our	-PRON-	PRP$
2163	127	3	example	example	NN
2163	127	4	the	the	DT
2163	127	5	dictionary	dictionary	JJ
2163	127	6	feature	feature	NN
2163	127	7	set	set	VBN
2163	127	8	would	would	MD
2163	127	9	be	be	VB
2163	127	10	:	:	:
2163	127	11	no	no	DT
2163	127	12	LastName	lastname	NN
2163	127	13	no	no	UH
2163	127	14	no	no	UH
2163	127	15	no	no	UH
2163	127	16	no	no	UH
2163	127	17	no	no	UH
2163	127	18	no	no	UH
2163	127	19	no	no	UH
2163	127	20	no	no	UH
2163	127	21			NN
2163	127	22	Date	date	NN
2163	127	23	check	check	NN
2163	127	24	—	—	:
2163	127	25	a	a	DT
2163	127	26	flag	flag	NN
2163	127	27	checking	checking	NN
2163	127	28	whether	whether	IN
2163	127	29	the	the	DT
2163	127	30	token	token	NN
2163	127	31	may	may	MD
2163	127	32	contain	contain	VB
2163	127	33	a	a	DT
2163	127	34	date	date	NN
2163	127	35	in	in	IN
2163	127	36	form	form	NN
2163	127	37	of	of	IN
2163	127	38	a	a	DT
2163	127	39	period	period	NN
2163	127	40	of	of	IN
2163	127	41	days	day	NNS
2163	127	42	,	,	,
2163	127	43	e.g.	e.g.	RB
2163	127	44	,	,	,
2163	127	45	12	12	CD
2163	127	46	-	-	SYM
2163	127	47	14	14	CD
2163	127	48	(	(	-LRB-
2163	127	49	possible	possible	JJ
2163	127	50	values	value	NNS
2163	127	51	:	:	:
2163	127	52	possDate	possdate	NN
2163	127	53	,	,	,
2163	127	54	no	no	UH
2163	127	55	)	)	-RRB-
2163	127	56	:	:	:
2163	127	57	no	no	DT
2163	127	58			:
2163	127	59	Pages	page	NNS
2163	127	60	check	check	VBP
2163	127	61	—	—	:
2163	127	62	a	a	DT
2163	127	63	flag	flag	NN
2163	127	64	checking	checking	NN
2163	127	65	whether	whether	IN
2163	127	66	the	the	DT
2163	127	67	token	token	NN
2163	127	68	may	may	MD
2163	127	69	contain	contain	VB
2163	127	70	pages	page	NNS
2163	127	71	,	,	,
2163	127	72	e.g.	e.g.	RB
2163	127	73	,	,	,
2163	127	74	234–238	234–238	CD
2163	127	75	(	(	-LRB-
2163	127	76	possible	possible	JJ
2163	127	77	values	value	NNS
2163	127	78	:	:	:
2163	127	79	possPages	posspage	NNS
2163	127	80	,	,	,
2163	127	81	no	no	UH
2163	127	82	)	)	-RRB-
2163	127	83	:	:	:
2163	127	84	no	no	DT
2163	127	85			JJ
2163	127	86	Token	token	JJ
2163	127	87	placement	placement	NN
2163	127	88	—	—	:
2163	127	89	the	the	DT
2163	127	90	token	token	JJ
2163	127	91	placement	placement	NN
2163	127	92	in	in	IN
2163	127	93	the	the	DT
2163	127	94	reference	reference	NN
2163	127	95	string	string	NN
2163	127	96	,	,	,
2163	127	97	based	base	VBN
2163	127	98	on	on	IN
2163	127	99	its	-PRON-	PRP$
2163	127	100	division	division	NN
2163	127	101	into	into	IN
2163	127	102	nine	nine	CD
2163	127	103	equal	equal	JJ
2163	127	104	consecutive	consecutive	JJ
2163	127	105	buckets	bucket	NNS
2163	127	106	.	.	.
2163	128	1	This	this	DT
2163	128	2	feature	feature	NN
2163	128	3	indicates	indicate	VBZ
2163	128	4	the	the	DT
2163	128	5	bucket	bucket	NN
2163	128	6	number	number	NN
2163	128	7	:	:	:
2163	128	8	0	0	NFP
2163	128	9	For	for	IN
2163	128	10	training	training	NN
2163	128	11	purposes	purpose	NNS
2163	128	12	we	-PRON-	PRP
2163	128	13	compiled	compile	VBD
2163	128	14	and	and	CC
2163	128	15	manually	manually	RB
2163	128	16	tagged	tag	VBD
2163	128	17	a	a	DT
2163	128	18	set	set	NN
2163	128	19	of	of	IN
2163	128	20	830	830	CD
2163	128	21	randomly	randomly	RB
2163	128	22	chosen	choose	VBN
2163	128	23	references	reference	NNS
2163	128	24	.	.	.
2163	129	1	These	these	DT
2163	129	2	were	be	VBD
2163	129	3	extracted	extract	VBN
2163	129	4	from	from	IN
2163	129	5	random	random	JJ
2163	129	6	publications	publication	NNS
2163	129	7	from	from	IN
2163	129	8	diverse	diverse	JJ
2163	129	9	conferences	conference	NNS
2163	129	10	and	and	CC
2163	129	11	journals	journal	NNS
2163	129	12	from	from	IN
2163	129	13	the	the	DT
2163	129	14	computer	computer	NN
2163	129	15	science	science	NN
2163	129	16	field	field	NN
2163	129	17	(	(	-LRB-
2163	129	18	collected	collect	VBN
2163	129	19	from	from	IN
2163	129	20	IEEE	IEEE	NNP
2163	129	21	Explorer	Explorer	NNP
2163	129	22	,	,	,
2163	129	23	Springer	Springer	NNP
2163	129	24	Link	Link	NNP
2163	129	25	or	or	CC
2163	129	26	the	the	DT
2163	129	27	ACM	ACM	NNP
2163	129	28	Portal	Portal	NNP
2163	129	29	)	)	-RRB-
2163	129	30	,	,	,
2163	129	31	manually	manually	RB
2163	129	32	cleaned	clean	VBN
2163	129	33	,	,	,
2163	129	34	tagged	tag	VBN
2163	129	35	,	,	,
2163	129	36	and	and	CC
2163	129	37	categorized	categorize	VBD
2163	129	38	according	accord	VBG
2163	129	39	to	to	IN
2163	129	40	their	-PRON-	PRP$
2163	129	41	type	type	NN
2163	129	42	of	of	IN
2163	129	43	publication	publication	NN
2163	129	44	venue	venue	NN
2163	129	45	.	.	.
2163	130	1	34	34	CD
2163	130	2	To	to	TO
2163	130	3	achieve	achieve	VB
2163	130	4	an	an	DT
2163	130	5	increased	increase	VBN
2163	130	6	versatility	versatility	NN
2163	130	7	,	,	,
2163	130	8	instead	instead	RB
2163	130	9	of	of	IN
2163	130	10	performing	perform	VBG
2163	130	11	cross-	cross-	JJ
2163	130	12	validation	validation	NN
2163	130	13	,	,	,
2163	130	14	35	35	CD
2163	130	15	which	which	WDT
2163	130	16	would	would	MD
2163	130	17	result	result	VB
2163	130	18	in	in	IN
2163	130	19	a	a	DT
2163	130	20	dataset-	dataset-	JJ
2163	130	21	tailored	tailored	JJ
2163	130	22	model	model	NN
2163	130	23	with	with	IN
2163	130	24	limited	limited	JJ
2163	130	25	or	or	CC
2163	130	26	no	no	DT
2163	130	27	versatility	versatility	NN
2163	130	28	,	,	,
2163	130	29	we	-PRON-	PRP
2163	130	30	opted	opt	VBD
2163	130	31	for	for	IN
2163	130	32	sampling	sample	VBG
2163	130	33	the	the	DT
2163	130	34	test	test	NN
2163	130	35	data	datum	NNS
2163	130	36	.	.	.
2163	131	1	Hence	hence	RB
2163	131	2	,	,	,
2163	131	3	we	-PRON-	PRP
2163	131	4	included	include	VBD
2163	131	5	in	in	IN
2163	131	6	the	the	DT
2163	131	7	training	training	NN
2163	131	8	corpus	corpus	NN
2163	131	9	some	some	DT
2163	131	10	samples	sample	NNS
2163	131	11	from	from	IN
2163	131	12	the	the	DT
2163	131	13	testing	testing	NN
2163	131	14	datasets	dataset	NNS
2163	131	15	as	as	IN
2163	131	16	follows	follow	VBZ
2163	131	17	:	:	:
2163	131	18	10	10	CD
2163	131	19	percent	percent	NN
2163	131	20	of	of	IN
2163	131	21	the	the	DT
2163	131	22	CORA	CORA	NNP
2163	131	23	dataset	dataset	NN
2163	131	24	(	(	-LRB-
2163	131	25	i.e.	i.e.	FW
2163	131	26	,	,	,
2163	131	27	20	20	CD
2163	131	28	entries	entry	NNS
2163	131	29	)	)	-RRB-
2163	131	30	,	,	,
2163	131	31	36	36	CD
2163	131	32	10	10	CD
2163	131	33	percent	percent	NN
2163	131	34	of	of	IN
2163	131	35	the	the	DT
2163	131	36	FLUX	FLUX	NNP
2163	131	37	-	-	HYPH
2163	131	38	CIM	CIM	NNP
2163	131	39	CS	CS	NNP
2163	131	40	dataset	dataset	NN
2163	131	41	(	(	-LRB-
2163	131	42	i.e.	i.e.	FW
2163	131	43	,	,	,
2163	131	44	30	30	CD
2163	131	45	entries	entry	NNS
2163	131	46	)	)	-RRB-
2163	131	47	,	,	,
2163	131	48	37	37	CD
2163	131	49	and	and	CC
2163	131	50	1	1	CD
2163	131	51	%	%	NN
2163	131	52	of	of	IN
2163	131	53	the	the	DT
2163	131	54	FLUX	FLUX	NNP
2163	131	55	-	-	HYPH
2163	131	56	CIM	CIM	NNP
2163	131	57	HS	HS	NNP
2163	131	58	dataset	dataset	NN
2163	131	59	(	(	-LRB-
2163	131	60	i.e.	i.e.	FW
2163	131	61	,	,	,
2163	131	62	20	20	CD
2163	131	63	entries	entry	NNS
2163	131	64	)	)	-RRB-
2163	131	65	.	.	.
2163	132	1	Consequently	consequently	RB
2163	132	2	,	,	,
2163	132	3	the	the	DT
2163	132	4	final	final	JJ
2163	132	5	training	training	NN
2163	132	6	corpus	corpus	NNP
2163	132	7	consisted	consist	VBD
2163	132	8	of	of	IN
2163	132	9	a	a	DT
2163	132	10	total	total	NN
2163	132	11	of	of	IN
2163	132	12	900	900	CD
2163	132	13	reference	reference	NN
2163	132	14	strings	string	NNS
2163	132	15	.	.	.
2163	133	1	To	to	TO
2163	133	2	clarify	clarify	VB
2163	133	3	,	,	,
2163	133	4	this	this	DT
2163	133	5	is	be	VBZ
2163	133	6	,	,	,
2163	133	7	to	to	IN
2163	133	8	some	some	DT
2163	133	9	extent	extent	NN
2163	133	10	,	,	,
2163	133	11	similar	similar	JJ
2163	133	12	to	to	IN
2163	133	13	the	the	DT
2163	133	14	dataset	dataset	NN
2163	133	15	-	-	HYPH
2163	133	16	specific	specific	JJ
2163	133	17	cross	cross	NN
2163	133	18	-	-	NN
2163	133	19	validation	validation	NN
2163	133	20	,	,	,
2163	133	21	but	but	CC
2163	133	22	instead	instead	RB
2163	133	23	of	of	IN
2163	133	24	considering	consider	VBG
2163	133	25	,	,	,
2163	133	26	for	for	IN
2163	133	27	example	example	NN
2163	133	28	,	,	,
2163	133	29	a	a	DT
2163	133	30	60–40	60–40	CD
2163	133	31	ratio	ratio	NN
2163	133	32	for	for	IN
2163	133	33	training	training	NN
2163	133	34	/	/	SYM
2163	133	35	testing	testing	NN
2163	133	36	,	,	,
2163	133	37	we	-PRON-	PRP
2163	133	38	used	use	VBD
2163	133	39	only	only	RB
2163	133	40	10	10	CD
2163	133	41	percent	percent	NN
2163	133	42	for	for	IN
2163	133	43	training	training	NN
2163	133	44	,	,	,
2163	133	45	while	while	IN
2163	133	46	the	the	DT
2163	133	47	testing	testing	NN
2163	133	48	(	(	-LRB-
2163	133	49	described	describe	VBN
2163	133	50	in	in	IN
2163	133	51	section	section	NN
2163	133	52	4	4	CD
2163	133	53	)	)	-RRB-
2163	133	54	was	be	VBD
2163	133	55	performed	perform	VBN
2163	133	56	as	as	IN
2163	133	57	a	a	DT
2163	133	58	direct	direct	JJ
2163	133	59	application	application	NN
2163	133	60	of	of	IN
2163	133	61	the	the	DT
2163	133	62	chunker	chunker	NN
2163	133	63	on	on	IN
2163	133	64	the	the	DT
2163	133	65	entire	entire	JJ
2163	133	66	dataset	dataset	NN
2163	133	67	.	.	.
2163	134	1	As	as	IN
2163	134	2	already	already	RB
2163	134	3	mentioned	mention	VBN
2163	134	4	,	,	,
2163	134	5	our	-PRON-	PRP$
2163	134	6	focus	focus	NN
2163	134	7	on	on	IN
2163	134	8	computer	computer	NN
2163	134	9	science	science	NN
2163	134	10	and	and	CC
2163	134	11	health	health	NN
2163	134	12	sciences	science	NNS
2163	134	13	is	be	VBZ
2163	134	14	strictly	strictly	RB
2163	134	15	due	due	JJ
2163	134	16	to	to	IN
2163	134	17	evaluation	evaluation	NN
2163	134	18	purposes	purpose	NNS
2163	134	19	.	.	.
2163	135	1	Our	-PRON-	PRP$
2163	135	2	proposed	propose	VBN
2163	135	3	model	model	NN
2163	135	4	is	be	VBZ
2163	135	5	domain	domain	NN
2163	135	6	-	-	HYPH
2163	135	7	agnostic	agnostic	JJ
2163	135	8	,	,	,
2163	135	9	and	and	CC
2163	135	10	hence	hence	RB
2163	135	11	,	,	,
2163	135	12	the	the	DT
2163	135	13	steps	step	NNS
2163	135	14	described	describe	VBN
2163	135	15	here	here	RB
2163	135	16	can	can	MD
2163	135	17	be	be	VB
2163	135	18	easily	easily	RB
2163	135	19	performed	perform	VBN
2163	135	20	on	on	IN
2163	135	21	datasets	dataset	NNS
2163	135	22	emerged	emerge	VBD
2163	135	23	from	from	IN
2163	135	24	other	other	JJ
2163	135	25	domains	domain	NNS
2163	135	26	,	,	,
2163	135	27	if	if	IN
2163	135	28	at	at	RB
2163	135	29	all	all	RB
2163	135	30	necessary	necessary	JJ
2163	135	31	.	.	.
2163	136	1	In	in	IN
2163	136	2	reality	reality	NN
2163	136	3	,	,	,
2163	136	4	the	the	DT
2163	136	5	chunker	chunker	NN
2163	136	6	’s	’s	POS
2163	136	7	performance	performance	NN
2163	136	8	on	on	IN
2163	136	9	references	reference	NNS
2163	136	10	from	from	IN
2163	136	11	a	a	DT
2163	136	12	domain	domain	NN
2163	136	13	not	not	RB
2163	136	14	covered	cover	VBN
2163	136	15	above	above	RB
2163	136	16	can	can	MD
2163	136	17	be	be	VB
2163	136	18	easily	easily	RB
2163	136	19	boosted	boost	VBN
2163	136	20	simply	simply	RB
2163	136	21	by	by	IN
2163	136	22	including	include	VBG
2163	136	23	a	a	DT
2163	136	24	sample	sample	NN
2163	136	25	of	of	IN
2163	136	26	references	reference	NNS
2163	136	27	in	in	IN
2163	136	28	the	the	DT
2163	136	29	training	training	NN
2163	136	30	set	set	VBN
2163	136	31	and	and	CC
2163	136	32	then	then	RB
2163	136	33	retraining	retrain	VBG
2163	136	34	the	the	DT
2163	136	35	chunker	chunker	NN
2163	136	36	.	.	.
2163	137	1	The	the	DT
2163	137	2	list	list	NN
2163	137	3	of	of	IN
2163	137	4	labels	label	NNS
2163	137	5	used	use	VBN
2163	137	6	for	for	IN
2163	137	7	training	training	NN
2163	137	8	and	and	CC
2163	137	9	then	then	RB
2163	137	10	testing	test	VBG
2163	137	11	consists	consist	NNS
2163	137	12	of	of	IN
2163	137	13	Author	author	NN
2163	137	14	,	,	,
2163	137	15	Title	Title	NNP
2163	137	16	,	,	,
2163	137	17	Journal	Journal	NNP
2163	137	18	,	,	,
2163	137	19	Conference	Conference	NNP
2163	137	20	,	,	,
2163	137	21	Workshop	Workshop	NNP
2163	137	22	,	,	,
2163	137	23	Website	Website	NNP
2163	137	24	,	,	,
2163	137	25	Technicalrep	Technicalrep	NNP
2163	137	26	,	,	,
2163	137	27	Date	Date	NNP
2163	137	28	,	,	,
2163	137	29	Publisher	publisher	NN
2163	137	30	,	,	,
2163	137	31	Location	Location	NNP
2163	137	32	,	,	,
2163	137	33	Volnum	Volnum	NNP
2163	137	34	,	,	,
2163	137	35	Pages	Pages	NNPS
2163	137	36	,	,	,
2163	137	37	Etal	Etal	NNP
2163	137	38	,	,	,
2163	137	39	Note	Note	NNP
2163	137	40	,	,	,
2163	137	41	Editors	Editors	NNP
2163	137	42	,	,	,
2163	137	43	Organization	Organization	NNP
2163	137	44	.	.	.
2163	138	1	As	as	IN
2163	138	2	we	-PRON-	PRP
2163	138	3	will	will	MD
2163	138	4	see	see	VB
2163	138	5	in	in	IN
2163	138	6	the	the	DT
2163	138	7	evaluation	evaluation	NN
2163	138	8	,	,	,
2163	138	9	not	not	RB
2163	138	10	all	all	DT
2163	138	11	labels	label	NNS
2163	138	12	were	be	VBD
2163	138	13	actually	actually	RB
2163	138	14	used	use	VBN
2163	138	15	for	for	IN
2163	138	16	testing	testing	NN
2163	138	17	(	(	-LRB-
2163	138	18	e.g.	e.g.	RB
2163	138	19	,	,	,
2163	138	20	Note	Note	NNP
2163	138	21	or	or	CC
2163	138	22	Editors	Editors	NNP
2163	138	23	)	)	-RRB-
2163	138	24	,	,	,
2163	138	25	some	some	DT
2163	138	26	of	of	IN
2163	138	27	them	-PRON-	PRP
2163	138	28	being	be	VBG
2163	138	29	present	present	JJ
2163	138	30	in	in	IN
2163	138	31	the	the	DT
2163	138	32	model	model	NN
2163	138	33	for	for	IN
2163	138	34	the	the	DT
2163	138	35	sake	sake	NN
2163	138	36	of	of	IN
2163	138	37	disambiguation	disambiguation	NN
2163	138	38	.	.	.
2163	139	1	Also	also	RB
2163	139	2	,	,	,
2163	139	3	as	as	IN
2163	139	4	opposed	oppose	VBN
2163	139	5	to	to	IN
2163	139	6	the	the	DT
2163	139	7	other	other	JJ
2163	139	8	approaches	approach	NNS
2163	139	9	,	,	,
2163	139	10	we	-PRON-	PRP
2163	139	11	made	make	VBD
2163	139	12	a	a	DT
2163	139	13	clear	clear	JJ
2163	139	14	distinction	distinction	NN
2163	139	15	between	between	IN
2163	139	16	Workshop	Workshop	NNP
2163	139	17	and	and	CC
2163	139	18	Conference	Conference	NNP
2163	139	19	,	,	,
2163	139	20	which	which	WDT
2163	139	21	adds	add	VBZ
2163	139	22	an	an	DT
2163	139	23	extra	extra	JJ
2163	139	24	degree	degree	NN
2163	139	25	to	to	IN
2163	139	26	the	the	DT
2163	139	27	complexity	complexity	NN
2163	139	28	of	of	IN
2163	139	29	the	the	DT
2163	139	30	disambiguation	disambiguation	NN
2163	139	31	.	.	.
2163	140	1	The	the	DT
2163	140	2	CRF	CRF	NNP
2163	140	3	model	model	NN
2163	140	4	was	be	VBD
2163	140	5	trained	train	VBN
2163	140	6	using	use	VBG
2163	140	7	the	the	DT
2163	140	8	MALLET	MALLET	NNP
2163	140	9	(	(	-LRB-
2163	140	10	A	a	DT
2163	140	11	Machine	Machine	NNP
2163	140	12	Learning	Learning	NNP
2163	140	13	for	for	IN
2163	140	14	Language	Language	NNP
2163	140	15	Toolkit	Toolkit	NNP
2163	140	16	)	)	-RRB-
2163	140	17	implementation	implementation	NN
2163	140	18	.	.	.
2163	141	1	38	38	CD
2163	141	2	The	the	DT
2163	141	3	output	output	NN
2163	141	4	of	of	IN
2163	141	5	the	the	DT
2163	141	6	chunker	chunker	NN
2163	141	7	is	be	VBZ
2163	141	8	post	post	JJ
2163	141	9	-	-	JJ
2163	141	10	processed	process	VBN
2163	141	11	to	to	TO
2163	141	12	expose	expose	VB
2163	141	13	a	a	DT
2163	141	14	series	series	NN
2163	141	15	of	of	IN
2163	141	16	fine	fine	RB
2163	141	17	-	-	HYPH
2163	141	18	grained	grain	VBN
2163	141	19	details	detail	NNS
2163	141	20	.	.	.
2163	142	1	As	as	IN
2163	142	2	shown	show	VBN
2163	142	3	in	in	IN
2163	142	4	figure	figure	NN
2163	142	5	1	1	CD
2163	142	6	in	in	IN
2163	142	7	all	all	PDT
2163	142	8	the	the	DT
2163	142	9	examples	example	NNS
2163	142	10	,	,	,
2163	142	11	the	the	DT
2163	142	12	chunking	chunking	NN
2163	142	13	provides	provide	VBZ
2163	142	14	a	a	DT
2163	142	15	blocked	block	VBN
2163	142	16	partition	partition	NN
2163	142	17	of	of	IN
2163	142	18	the	the	DT
2163	142	19	reference	reference	NN
2163	142	20	string	string	NN
2163	142	21	,	,	,
2163	142	22	but	but	CC
2163	142	23	we	-PRON-	PRP
2163	142	24	require	require	VBP
2163	142	25	for	for	IN
2163	142	26	the	the	DT
2163	142	27	Author	author	NN
2163	142	28	field	field	NN
2163	142	29	an	an	DT
2163	142	30	even	even	RB
2163	142	31	deeper	deep	JJR
2163	142	32	partition	partition	NN
2163	142	33	.	.	.
2163	143	1	Consequently	consequently	RB
2163	143	2	,	,	,
2163	143	3	following	follow	VBG
2163	143	4	a	a	DT
2163	143	5	rule	rule	NN
2163	143	6	-	-	HYPH
2163	143	7	based	base	VBN
2163	143	8	approach	approach	NN
2163	143	9	we	-PRON-	PRP
2163	143	10	extract	extract	VBP
2163	143	11	the	the	DT
2163	143	12	individual	individual	JJ
2163	143	13	author	author	NN
2163	143	14	names	name	NNS
2163	143	15	from	from	IN
2163	143	16	the	the	DT
2163	143	17	Author	author	NN
2163	143	18	block	block	NN
2163	143	19	making	make	VBG
2163	143	20	use	use	NN
2163	143	21	of	of	IN
2163	143	22	the	the	DT
2163	143	23	punctuation	punctuation	NN
2163	143	24	marks	mark	NNS
2163	143	25	,	,	,
2163	143	26	the	the	DT
2163	143	27	orthographic	orthographic	JJ
2163	143	28	case	case	NN
2163	143	29	,	,	,
2163	143	30	and	and	CC
2163	143	31	the	the	DT
2163	143	32	alternation	alternation	NN
2163	143	33	between	between	IN
2163	143	34	initials	initial	NNS
2163	143	35	and	and	CC
2163	143	36	actual	actual	JJ
2163	143	37	names	name	NNS
2163	143	38	.	.	.
2163	144	1	When	when	WRB
2163	144	2	no	no	DT
2163	144	3	initials	initial	NNS
2163	144	4	,	,	,
2163	144	5	subject	subject	JJ
2163	144	6	to	to	IN
2163	144	7	the	the	DT
2163	144	8	existing	exist	VBG
2163	144	9	punctuation	punctuation	NN
2163	144	10	marks	mark	NNS
2163	144	11	,	,	,
2163	144	12	we	-PRON-	PRP
2163	144	13	consider	consider	VBP
2163	144	14	as	as	IN
2163	144	15	a	a	DT
2163	144	16	rule	rule	NN
2163	144	17	-	-	HYPH
2163	144	18	of	of	IN
2163	144	19	-	-	HYPH
2163	144	20	thumb	thumb	NN
2163	144	21	that	that	WDT
2163	144	22	each	each	DT
2163	144	23	name	name	NN
2163	144	24	generally	generally	RB
2163	144	25	comprises	comprise	VBZ
2163	144	26	one	one	CD
2163	144	27	first	first	JJ
2163	144	28	name	name	NN
2163	144	29	and	and	CC
2163	144	30	one	one	CD
2163	144	31	surname	surname	NN
2163	144	32	(	(	-LRB-
2163	144	33	in	in	IN
2163	144	34	this	this	DT
2163	144	35	order	order	NN
2163	144	36	,	,	,
2163	144	37	i.e.	i.e.	FW
2163	144	38	,	,	,
2163	144	39	John	John	NNP
2163	144	40	Doe	Doe	NNP
2163	144	41	)	)	-RRB-
2163	144	42	.	.	.
2163	145	1	The	the	DT
2163	145	2	result	result	NN
2163	145	3	of	of	IN
2163	145	4	the	the	DT
2163	145	5	post	post	NN
2163	145	6	-	-	JJ
2163	145	7	processing	processing	NN
2163	145	8	is	be	VBZ
2163	145	9	used	use	VBN
2163	145	10	in	in	IN
2163	145	11	the	the	DT
2163	145	12	linking	linking	NN
2163	145	13	process	process	NN
2163	145	14	.	.	.
2163	146	1	REFERENCE	reference	NN
2163	146	2	INFORMATION	INFORMATION	NNS
2163	146	3	EXTRACTION	extraction	NN
2163	146	4	AND	and	CC
2163	146	5	PROCESSING	processing	NN
2163	146	6	|GROZA	|groza	NN
2163	146	7	,	,	,
2163	146	8	GRIMNES	GRIMNES	NNP
2163	146	9	,	,	,
2163	146	10	AND	and	CC
2163	146	11	HANDSCHUH	handschuh	NN
2163	146	12	14	14	CD
2163	146	13	4	4	CD
2163	146	14	.	.	.
2163	147	1	EXPERIMENTAL	experimental	NN
2163	147	2	RESULTS	results	VBP
2163	147	3	We	-PRON-	PRP
2163	147	4	have	have	VBP
2163	147	5	performed	perform	VBN
2163	147	6	an	an	DT
2163	147	7	extensive	extensive	JJ
2163	147	8	evaluation	evaluation	NN
2163	147	9	of	of	IN
2163	147	10	the	the	DT
2163	147	11	proposed	propose	VBN
2163	147	12	reference	reference	NN
2163	147	13	chunking	chunk	VBG
2163	147	14	approach	approach	NN
2163	147	15	.	.	.
2163	148	1	In	in	IN
2163	148	2	general	general	JJ
2163	148	3	,	,	,
2163	148	4	all	all	PDT
2163	148	5	the	the	DT
2163	148	6	previous	previous	JJ
2163	148	7	work	work	NN
2163	148	8	in	in	IN
2163	148	9	reference	reference	NN
2163	148	10	chunking	chunk	VBG
2163	148	11	focuses	focus	NNS
2163	148	12	on	on	IN
2163	148	13	raw	raw	JJ
2163	148	14	reference	reference	NN
2163	148	15	chunking	chunking	NN
2163	148	16	,	,	,
2163	148	17	i.e.	i.e.	FW
2163	148	18	,	,	,
2163	148	19	label	label	NNP
2163	148	20	sequencing	sequence	VBG
2163	148	21	at	at	IN
2163	148	22	the	the	DT
2163	148	23	macro	macro	JJ
2163	148	24	level	level	NN
2163	148	25	.	.	.
2163	149	1	More	more	RBR
2163	149	2	concretely	concretely	JJ
2163	149	3	,	,	,
2163	149	4	the	the	DT
2163	149	5	other	other	JJ
2163	149	6	approaches	approach	NNS
2163	149	7	split	split	VBP
2163	149	8	and	and	CC
2163	149	9	tag	tag	VB
2163	149	10	the	the	DT
2163	149	11	reference	reference	NN
2163	149	12	strings	string	NNS
2163	149	13	using	use	VBG
2163	149	14	blocks	block	NNS
2163	149	15	of	of	IN
2163	149	16	complete	complete	JJ
2163	149	17	references	reference	NNS
2163	149	18	,	,	,
2163	149	19	without	without	IN
2163	149	20	going	go	VBG
2163	149	21	into	into	IN
2163	149	22	details	detail	NNS
2163	149	23	such	such	JJ
2163	149	24	as	as	IN
2163	149	25	chunking	chunk	VBG
2163	149	26	individual	individual	JJ
2163	149	27	authors	author	NNS
2163	149	28	.	.	.
2163	150	1	The	the	DT
2163	150	2	only	only	JJ
2163	150	3	exception	exception	NN
2163	150	4	is	be	VBZ
2163	150	5	the	the	DT
2163	150	6	ParsCit	ParsCit	NNP
2163	150	7	package	package	NN
2163	150	8	that	that	WDT
2163	150	9	does	do	VBZ
2163	150	10	perform	perform	VB
2163	150	11	complete	complete	JJ
2163	150	12	reference	reference	NN
2163	150	13	chunking	chunk	VBG
2163	150	14	in	in	IN
2163	150	15	a	a	DT
2163	150	16	similar	similar	JJ
2163	150	17	fashion	fashion	NN
2163	150	18	as	as	IN
2163	150	19	we	-PRON-	PRP
2163	150	20	do	do	VBP
2163	150	21	.	.	.
2163	151	1	The	the	DT
2163	151	2	evaluation	evaluation	NN
2163	151	3	results	result	NNS
2163	151	4	presented	present	VBN
2163	151	5	in	in	IN
2163	151	6	this	this	DT
2163	151	7	section	section	NN
2163	151	8	,	,	,
2163	151	9	will	will	MD
2163	151	10	feature	feature	VB
2163	151	11	complete	complete	JJ
2163	151	12	chunking	chunk	VBG
2163	151	13	only	only	RB
2163	151	14	for	for	IN
2163	151	15	our	-PRON-	PRP$
2163	151	16	solution	solution	NN
2163	151	17	and	and	CC
2163	151	18	for	for	IN
2163	151	19	ParsCit	ParsCit	NNP
2163	151	20	,	,	,
2163	151	21	and	and	CC
2163	151	22	raw	raw	JJ
2163	151	23	chunking	chunk	VBG
2163	151	24	for	for	IN
2163	151	25	the	the	DT
2163	151	26	rest	rest	NN
2163	151	27	of	of	IN
2163	151	28	the	the	DT
2163	151	29	approaches	approach	NNS
2163	151	30	.	.	.
2163	152	1	Field	Field	NNP
2163	152	2	ParsCit	ParsCit	NNP
2163	152	3	Peng	Peng	NNP
2163	152	4	Han	Han	NNP
2163	152	5	et	et	NNP
2163	152	6	al	al	NNP
2163	152	7	.	.	.
2163	153	1	Our	-PRON-	PRP$
2163	153	2	approach	approach	NN
2163	153	3	P	P	NNP
2163	153	4	R	r	NN
2163	153	5	F1	f1	NN
2163	153	6	F1	f1	NN
2163	153	7	P	P	NNP
2163	153	8	R	R	NNP
2163	153	9	F1	f1	NN
2163	153	10	P	P	NNP
2163	153	11	R	R	NNP
2163	153	12	F1	F1	NNP
2163	153	13	Author	author	NN
2163	153	14	98.7	98.7	CD
2163	153	15	99.3	99.3	CD
2163	153	16	98.99	98.99	CD
2163	153	17	99.4	99.4	CD
2163	153	18	92.6	92.6	CD
2163	153	19	99.1	99.1	CD
2163	153	20	97.6	97.6	CD
2163	153	21	99.08	99.08	CD
2163	153	22	99.6	99.6	CD
2163	153	23	99.30	99.30	CD
2163	153	24	Title	Title	NNP
2163	153	25	96.0	96.0	CD
2163	153	26	98.4	98.4	CD
2163	153	27	97.18	97.18	CD
2163	153	28	98.3	98.3	CD
2163	153	29	92.2	92.2	CD
2163	153	30	93.0	93.0	CD
2163	153	31	92.6	92.6	CD
2163	153	32	95.64	95.64	CD
2163	153	33	95.64	95.64	CD
2163	153	34	95.64	95.64	CD
2163	153	35	Date	date	NN
2163	153	36	100	100	CD
2163	153	37	98.4	98.4	CD
2163	153	38	99.19	99.19	CD
2163	153	39	98.9	98.9	CD
2163	153	40	98.5	98.5	CD
2163	153	41	95.9	95.9	CD
2163	153	42	97.2	97.2	CD
2163	153	43	99.33	99.33	CD
2163	153	44	98.67	98.67	CD
2163	153	45	98.99	98.99	CD
2163	153	46	Pages	Pages	NNPS
2163	153	47	97.7	97.7	CD
2163	153	48	98.4	98.4	CD
2163	153	49	98.04	98.04	CD
2163	153	50	98.6	98.6	CD
2163	153	51	95.6	95.6	CD
2163	153	52	96.9	96.9	CD
2163	153	53	96.2	96.2	CD
2163	153	54	99.28	99.28	CD
2163	153	55	99.22	99.22	CD
2163	153	56	99.24	99.24	CD
2163	153	57	Location	Location	NNP
2163	153	58	95.6	95.6	CD
2163	153	59	90.0	90.0	CD
2163	153	60	92.71	92.71	CD
2163	153	61	87.2	87.2	CD
2163	153	62	77.7	77.7	CD
2163	153	63	71.5	71.5	CD
2163	153	64	74.5	74.5	CD
2163	153	65	93.45	93.45	CD
2163	153	66	92.59	92.59	CD
2163	153	67	93.01	93.01	CD
2163	153	68	Organization	Organization	NNP
2163	153	69	90.9	90.9	CD
2163	153	70	87.9	87.9	CD
2163	153	71	89.37	89.37	CD
2163	153	72	94.0	94.0	CD
2163	153	73	76.5	76.5	CD
2163	153	74	77.3	77.3	CD
2163	153	75	76.9	76.9	CD
2163	153	76	100	100	CD
2163	153	77	87.87	87.87	CD
2163	153	78	93.54	93.54	CD
2163	153	79	Journal	Journal	NNP
2163	153	80	90.8	90.8	CD
2163	153	81	91.2	91.2	CD
2163	153	82	90.99	90.99	CD
2163	153	83	91.3	91.3	CD
2163	153	84	77.1	77.1	CD
2163	153	85	78.7	78.7	CD
2163	153	86	77.9	77.9	CD
2163	153	87	94.02	94.02	CD
2163	153	88	97.42	97.42	CD
2163	153	89	95.68	95.68	CD
2163	153	90	Booktitle	Booktitle	NNP
2163	153	91	92.7	92.7	CD
2163	153	92	94.2	94.2	CD
2163	153	93	93.44	93.44	CD
2163	153	94	93.7	93.7	CD
2163	153	95	88.7	88.7	CD
2163	153	96	88.9	88.9	CD
2163	153	97	88.88	88.88	CD
2163	153	98	97.77	97.77	CD
2163	153	99	98.44	98.44	CD
2163	153	100	98.10	98.10	CD
2163	153	101	Publisher	Publisher	NNP
2163	153	102	95.2	95.2	CD
2163	153	103	88.7	88.7	CD
2163	153	104	91.83	91.83	CD
2163	153	105	76.1	76.1	CD
2163	153	106	56.0	56.0	CD
2163	153	107	64.1	64.1	CD
2163	153	108	59.9	59.9	CD
2163	153	109	94.84	94.84	CD
2163	153	110	95.83	95.83	CD
2163	153	111	95.33	95.33	CD
2163	153	112	Tech	Tech	NNP
2163	153	113	.	.	.
2163	154	1	rep	rep	NNP
2163	154	2	.	.	NNP
2163	154	3	94.0	94.0	CD
2163	154	4	79.6	79.6	CD
2163	154	5	86.2	86.2	CD
2163	154	6	86.7	86.7	CD
2163	154	7	56.2	56.2	CD
2163	154	8	64.1	64.1	CD
2163	154	9	59.9	59.9	CD
2163	154	10	100	100	CD
2163	154	11	90.90	90.90	CD
2163	154	12	95.23	95.23	CD
2163	154	13	Website	Website	NNP
2163	154	14	-	-	HYPH
2163	154	15	-	-	HYPH
2163	154	16	-	-	HYPH
2163	154	17	-	-	HYPH
2163	154	18	-	-	HYPH
2163	154	19	-	-	HYPH
2163	154	20	-	-	HYPH
2163	154	21	100	100	CD
2163	154	22	100	100	CD
2163	154	23	100	100	CD
2163	154	24	Table	table	NN
2163	154	25	1	1	CD
2163	154	26	.	.	.
2163	155	1	Evaluation	Evaluation	NNP
2163	155	2	Results	Results	NNP
2163	155	3	on	on	IN
2163	155	4	the	the	DT
2163	155	5	CORA	CORA	NNP
2163	155	6	Dataset	Dataset	NNP
2163	155	7	An	an	DT
2163	155	8	additional	additional	JJ
2163	155	9	observation	observation	NN
2163	155	10	we	-PRON-	PRP
2163	155	11	need	need	VBP
2163	155	12	to	to	TO
2163	155	13	make	make	VB
2163	155	14	is	be	VBZ
2163	155	15	related	relate	VBN
2163	155	16	to	to	IN
2163	155	17	the	the	DT
2163	155	18	reference	reference	NN
2163	155	19	fields	field	NNS
2163	155	20	taken	take	VBN
2163	155	21	into	into	IN
2163	155	22	account	account	NN
2163	155	23	.	.	.
2163	156	1	Most	Most	JJS
2163	156	2	of	of	IN
2163	156	3	the	the	DT
2163	156	4	fields	field	NNS
2163	156	5	we	-PRON-	PRP
2163	156	6	have	have	VBP
2163	156	7	focused	focus	VBN
2163	156	8	on	on	IN
2163	156	9	coincide	coincide	NN
2163	156	10	with	with	IN
2163	156	11	the	the	DT
2163	156	12	fields	field	NNS
2163	156	13	considered	consider	VBN
2163	156	14	by	by	IN
2163	156	15	all	all	PDT
2163	156	16	the	the	DT
2163	156	17	existing	exist	VBG
2163	156	18	relevant	relevant	JJ
2163	156	19	approaches	approach	NNS
2163	156	20	.	.	.
2163	157	1	Nevertheless	nevertheless	RB
2163	157	2	,	,	,
2163	157	3	there	there	EX
2163	157	4	are	be	VBP
2163	157	5	also	also	RB
2163	157	6	some	some	DT
2163	157	7	discrepancies	discrepancy	NNS
2163	157	8	,	,	,
2163	157	9	listed	list	VBN
2163	157	10	as	as	IN
2163	157	11	follows	follow	VBZ
2163	157	12	:	:	:
2163	157	13			NFP
2163	157	14	the	the	DT
2163	157	15	fields	field	NNS
2163	157	16	:	:	:
2163	157	17	Volume	volume	NN
2163	157	18	,	,	,
2163	157	19	Number	Number	NNP
2163	157	20	,	,	,
2163	157	21	Editors	Editors	NNP
2163	157	22	,	,	,
2163	157	23	or	or	CC
2163	157	24	Note	Note	NNP
2163	157	25	were	be	VBD
2163	157	26	used	use	VBN
2163	157	27	in	in	IN
2163	157	28	the	the	DT
2163	157	29	chunking	chunking	NN
2163	157	30	process	process	NN
2163	157	31	b	b	NNP
2163	157	32	u	u	NNP
2163	157	33	t	t	NNP
2163	157	34	are	be	VBP
2163	157	35	not	not	RB
2163	157	36	considered	consider	VBN
2163	157	37	for	for	IN
2163	157	38	evaluation	evaluation	NN
2163	157	39			NN
2163	157	40	unlike	unlike	IN
2163	157	41	all	all	PDT
2163	157	42	the	the	DT
2163	157	43	other	other	JJ
2163	157	44	approaches	approach	NNS
2163	157	45	,	,	,
2163	157	46	we	-PRON-	PRP
2163	157	47	make	make	VBP
2163	157	48	the	the	DT
2163	157	49	distinction	distinction	NN
2163	157	50	between	between	IN
2163	157	51	Conference	Conference	NNP
2163	157	52	and	and	CC
2163	157	53	Workshop	Workshop	NNP
2163	157	54	as	as	IN
2163	157	55	publication	publication	NN
2163	157	56	venues	venue	NNS
2163	157	57	.	.	.
2163	158	1	However	however	RB
2163	158	2	,	,	,
2163	158	3	for	for	IN
2163	158	4	alignment	alignment	NN
2163	158	5	purposes	purpose	NNS
2163	158	6	(	(	-LRB-
2163	158	7	i.e.	i.e.	FW
2163	158	8	,	,	,
2163	158	9	to	to	TO
2163	158	10	be	be	VB
2163	158	11	able	able	JJ
2163	158	12	to	to	TO
2163	158	13	compare	compare	VB
2163	158	14	our	-PRON-	PRP$
2163	158	15	results	result	NNS
2163	158	16	with	with	IN
2163	158	17	the	the	DT
2163	158	18	other	other	JJ
2163	158	19	approaches	approach	NNS
2163	158	20	)	)	-RRB-
2163	158	21	,	,	,
2163	158	22	in	in	IN
2163	158	23	the	the	DT
2163	158	24	evaluation	evaluation	NN
2163	158	25	results	result	NNS
2163	158	26	these	these	DT
2163	158	27	are	be	VBP
2163	158	28	merged	merge	VBN
2163	158	29	into	into	IN
2163	158	30	the	the	DT
2163	158	31	Booktitle	Booktitle	NNP
2163	158	32	field	field	NN
2163	158	33	.	.	.
2163	159	1	The	the	DT
2163	159	2	actual	actual	JJ
2163	159	3	tests	test	NNS
2163	159	4	were	be	VBD
2163	159	5	performed	perform	VBN
2163	159	6	on	on	IN
2163	159	7	four	four	CD
2163	159	8	different	different	JJ
2163	159	9	datasets	dataset	NNS
2163	159	10	,	,	,
2163	159	11	three	three	CD
2163	159	12	of	of	IN
2163	159	13	them	-PRON-	PRP
2163	159	14	used	use	VBN
2163	159	15	also	also	RB
2163	159	16	for	for	IN
2163	159	17	evaluating	evaluate	VBG
2163	159	18	the	the	DT
2163	159	19	other	other	JJ
2163	159	20	approaches	approach	NNS
2163	159	21	,	,	,
2163	159	22	and	and	CC
2163	159	23	a	a	DT
2163	159	24	fourth	fourth	JJ
2163	159	25	one	one	NN
2163	159	26	compiled	compile	VBN
2163	159	27	by	by	IN
2163	159	28	us	-PRON-	PRP
2163	159	29	.	.	.
2163	160	1	In	in	IN
2163	160	2	the	the	DT
2163	160	3	case	case	NN
2163	160	4	of	of	IN
2163	160	5	the	the	DT
2163	160	6	three	three	CD
2163	160	7	existing	exist	VBG
2163	160	8	datasets	dataset	NNS
2163	160	9	,	,	,
2163	160	10	during	during	IN
2163	160	11	the	the	DT
2163	160	12	experimental	experimental	JJ
2163	160	13	evaluation	evaluation	NN
2163	160	14	we	-PRON-	PRP
2163	160	15	did	do	VBD
2163	160	16	not	not	RB
2163	160	17	make	make	VB
2163	160	18	use	use	NN
2163	160	19	of	of	IN
2163	160	20	the	the	DT
2163	160	21	preprocessing	preprocesse	VBG
2163	160	22	step	step	NN
2163	160	23	as	as	IN
2163	160	24	they	-PRON-	PRP
2163	160	25	were	be	VBD
2163	160	26	already	already	RB
2163	160	27	clean	clean	JJ
2163	160	28	.	.	.
2163	161	1	As	as	IN
2163	161	2	evaluation	evaluation	NN
2163	161	3	metric	metric	NN
2163	161	4	,	,	,
2163	161	5	we	-PRON-	PRP
2163	161	6	used	use	VBD
2163	161	7	the	the	DT
2163	161	8	F1	F1	NNP
2163	161	9	score	score	NN
2163	161	10	,	,	,
2163	161	11	39	39	CD
2163	161	12	i.e.	i.e.	FW
2163	161	13	,	,	,
2163	161	14	the	the	DT
2163	161	15	harmonic	harmonic	JJ
2163	161	16	mean	mean	NN
2163	161	17	of	of	IN
2163	161	18	precision	precision	NNP
2163	161	19	and	and	CC
2163	161	20	recall	recall	NN
2163	161	21	,	,	,
2163	161	22	using	use	VBG
2163	161	23	the	the	DT
2163	161	24	following	follow	VBG
2163	161	25	formula	formula	NN
2163	161	26	:	:	:
2163	161	27	INFORMATION	INFORMATION	NNP
2163	161	28	TECHNOLOGY	TECHNOLOGY	NNP
2163	161	29	AND	and	CC
2163	161	30	LIBRARIES	LIBRARIES	NNP
2163	161	31	|	|	NNP
2163	161	32	JUNE	JUNE	NNP
2163	161	33	2012	2012	CD
2163	161	34	15	15	CD
2163	161	35	In	in	IN
2163	161	36	the	the	DT
2163	161	37	following	following	NN
2163	161	38	,	,	,
2163	161	39	we	-PRON-	PRP
2163	161	40	iterate	iterate	VBP
2163	161	41	over	over	IN
2163	161	42	each	each	DT
2163	161	43	dataset	dataset	NN
2163	161	44	,	,	,
2163	161	45	by	by	IN
2163	161	46	providing	provide	VBG
2163	161	47	a	a	DT
2163	161	48	short	short	JJ
2163	161	49	description	description	NN
2163	161	50	and	and	CC
2163	161	51	the	the	DT
2163	161	52	experimental	experimental	JJ
2163	161	53	results	result	NNS
2163	161	54	.	.	.
2163	162	1	It	-PRON-	PRP
2163	162	2	is	be	VBZ
2163	162	3	worth	worth	JJ
2163	162	4	mentioning	mention	VBG
2163	162	5	that	that	IN
2163	162	6	our	-PRON-	PRP$
2163	162	7	CRF	CRF	NNP
2163	162	8	reference	reference	NN
2163	162	9	chunker	chunker	NN
2163	162	10	was	be	VBD
2163	162	11	trained	train	VBN
2163	162	12	only	only	RB
2163	162	13	once	once	RB
2163	162	14	,	,	,
2163	162	15	as	as	IN
2163	162	16	described	describe	VBN
2163	162	17	earlier	early	RBR
2163	162	18	,	,	,
2163	162	19	and	and	CC
2163	162	20	not	not	RB
2163	162	21	specifically	specifically	RB
2163	162	22	for	for	IN
2163	162	23	each	each	DT
2163	162	24	dataset	dataset	NN
2163	162	25	.	.	.
2163	163	1	4.1	4.1	CD
2163	163	2	Dataset	dataset	NN
2163	163	3	:	:	:
2163	163	4	CORA	cora	VB
2163	163	5	The	the	DT
2163	163	6	CORA	CORA	NNP
2163	163	7	dataset	dataset	NN
2163	163	8	is	be	VBZ
2163	163	9	the	the	DT
2163	163	10	first	first	JJ
2163	163	11	gold	gold	NN
2163	163	12	standard	standard	NN
2163	163	13	created	create	VBN
2163	163	14	for	for	IN
2163	163	15	automatic	automatic	JJ
2163	163	16	reference	reference	NN
2163	163	17	chunking	chunk	VBG
2163	163	18	.	.	.
2163	164	1	40	40	CD
2163	164	2	It	-PRON-	PRP
2163	164	3	comprises	comprise	VBZ
2163	164	4	two	two	CD
2163	164	5	hundred	hundred	CD
2163	164	6	reference	reference	NN
2163	164	7	strings	string	NNS
2163	164	8	and	and	CC
2163	164	9	focuses	focus	VBZ
2163	164	10	on	on	IN
2163	164	11	the	the	DT
2163	164	12	computer	computer	NN
2163	164	13	science	science	NN
2163	164	14	area	area	NN
2163	164	15	.	.	.
2163	165	1	Each	each	DT
2163	165	2	entry	entry	NN
2163	165	3	is	be	VBZ
2163	165	4	segmented	segment	VBN
2163	165	5	into	into	IN
2163	165	6	thirteen	thirteen	CD
2163	165	7	different	different	JJ
2163	165	8	fields	field	NNS
2163	165	9	:	:	:
2163	165	10	Author	author	NN
2163	165	11	,	,	,
2163	165	12	Editor	Editor	NNP
2163	165	13	,	,	,
2163	165	14	Title	Title	NNP
2163	165	15	,	,	,
2163	165	16	Booktitle	Booktitle	NNP
2163	165	17	,	,	,
2163	165	18	Journal	Journal	NNP
2163	165	19	,	,	,
2163	165	20	Volume	volume	NN
2163	165	21	,	,	,
2163	165	22	Publisher	publisher	NN
2163	165	23	,	,	,
2163	165	24	Date	date	NN
2163	165	25	,	,	,
2163	165	26	Pages	Pages	NNPS
2163	165	27	,	,	,
2163	165	28	Location	Location	NNP
2163	165	29	,	,	,
2163	165	30	Tech	Tech	NNP
2163	165	31	,	,	,
2163	165	32	Institution	Institution	NNP
2163	165	33	and	and	CC
2163	165	34	Note	Note	NNP
2163	165	35	.	.	.
2163	166	1	Table	table	NN
2163	166	2	1	1	CD
2163	166	3	shows	show	VBZ
2163	166	4	the	the	DT
2163	166	5	comparative	comparative	JJ
2163	166	6	evaluation	evaluation	NN
2163	166	7	results	result	NNS
2163	166	8	on	on	IN
2163	166	9	the	the	DT
2163	166	10	CORA	CORA	NNP
2163	166	11	dataset	dataset	NN
2163	166	12	of	of	IN
2163	166	13	ParsCit	ParsCit	NNP
2163	166	14	,	,	,
2163	166	15	Peng	Peng	NNP
2163	166	16	et	et	NNP
2163	166	17	al	al	NNP
2163	166	18	.	.	NNP
2163	166	19	,	,	,
2163	166	20	41	41	CD
2163	166	21	Han	Han	NNP
2163	166	22	et	et	NNP
2163	166	23	al	al	NNP
2163	166	24	.	.	NNP
2163	166	25	,	,	,
2163	166	26	42	42	CD
2163	166	27	and	and	CC
2163	166	28	our	-PRON-	PRP$
2163	166	29	approach	approach	NN
2163	166	30	.	.	.
2163	167	1	We	-PRON-	PRP
2163	167	2	observe	observe	VBP
2163	167	3	that	that	IN
2163	167	4	our	-PRON-	PRP$
2163	167	5	chunker	chunker	NN
2163	167	6	outperforms	outperform	VBZ
2163	167	7	the	the	DT
2163	167	8	other	other	JJ
2163	167	9	chunkers	chunker	NNS
2163	167	10	on	on	IN
2163	167	11	most	most	JJS
2163	167	12	of	of	IN
2163	167	13	the	the	DT
2163	167	14	fields	field	NNS
2163	167	15	,	,	,
2163	167	16	with	with	IN
2163	167	17	some	some	DT
2163	167	18	of	of	IN
2163	167	19	them	-PRON-	PRP
2163	167	20	presenting	present	VBG
2163	167	21	a	a	DT
2163	167	22	significant	significant	JJ
2163	167	23	increase	increase	NN
2163	167	24	in	in	IN
2163	167	25	performance	performance	NN
2163	167	26	(	(	-LRB-
2163	167	27	looking	look	VBG
2163	167	28	at	at	IN
2163	167	29	the	the	DT
2163	167	30	F1	F1	NNP
2163	167	31	score	score	NN
2163	167	32	)	)	-RRB-
2163	167	33	:	:	:
2163	167	34	Journal	Journal	NNP
2163	167	35	from	from	IN
2163	167	36	91.3	91.3	CD
2163	167	37	percent	percent	NN
2163	167	38	to	to	IN
2163	167	39	95.68	95.68	CD
2163	167	40	percent	percent	NN
2163	167	41	,	,	,
2163	167	42	Booktitle	Booktitle	NNP
2163	167	43	from	from	IN
2163	167	44	93.44	93.44	CD
2163	167	45	percent	percent	NN
2163	167	46	to	to	IN
2163	167	47	98.10	98.10	CD
2163	167	48	percent	percent	NN
2163	167	49	,	,	,
2163	167	50	Publisher	publisher	NN
2163	167	51	from	from	IN
2163	167	52	91.83	91.83	CD
2163	167	53	percent	percent	NN
2163	167	54	to	to	IN
2163	167	55	95.33	95.33	CD
2163	167	56	percent	percent	NN
2163	167	57	,	,	,
2163	167	58	and	and	CC
2163	167	59	especially	especially	RB
2163	167	60	Tech	Tech	NNP
2163	167	61	.	.	.
2163	168	1	rep	rep	NNP
2163	168	2	.	.	NNP
2163	168	3	from	from	IN
2163	168	4	86.7	86.7	CD
2163	168	5	percent	percent	NN
2163	168	6	to	to	IN
2163	168	7	95.23	95.23	CD
2163	168	8	percent	percent	NN
2163	168	9	.	.	.
2163	169	1	In	in	IN
2163	169	2	the	the	DT
2163	169	3	case	case	NN
2163	169	4	of	of	IN
2163	169	5	the	the	DT
2163	169	6	fields	field	NNS
2163	169	7	where	where	WRB
2163	169	8	our	-PRON-	PRP$
2163	169	9	chunker	chunker	NN
2163	169	10	was	be	VBD
2163	169	11	outperformed	outperform	VBN
2163	169	12	,	,	,
2163	169	13	the	the	DT
2163	169	14	F1	f1	NN
2163	169	15	score	score	NN
2163	169	16	is	be	VBZ
2163	169	17	very	very	RB
2163	169	18	close	close	JJ
2163	169	19	to	to	IN
2163	169	20	the	the	DT
2163	169	21	best	good	JJS
2163	169	22	of	of	IN
2163	169	23	the	the	DT
2163	169	24	approaches	approach	NNS
2163	169	25	and	and	CC
2163	169	26	includes	include	VBZ
2163	169	27	an	an	DT
2163	169	28	increase	increase	NN
2163	169	29	in	in	IN
2163	169	30	one	one	CD
2163	169	31	of	of	IN
2163	169	32	its	-PRON-	PRP$
2163	169	33	two	two	CD
2163	169	34	components	component	NNS
2163	169	35	(	(	-LRB-
2163	169	36	i.e.	i.e.	FW
2163	169	37	,	,	,
2163	169	38	precision	precision	NN
2163	169	39	or	or	CC
2163	169	40	recall	recall	NN
2163	169	41	)	)	-RRB-
2163	169	42	.	.	.
2163	170	1	For	for	IN
2163	170	2	example	example	NN
2163	170	3	,	,	,
2163	170	4	on	on	IN
2163	170	5	the	the	DT
2163	170	6	Organization	Organization	NNP
2163	170	7	field	field	NN
2163	170	8	,	,	,
2163	170	9	we	-PRON-	PRP
2163	170	10	scored	score	VBD
2163	170	11	93.54percent	93.54percent	CD
2163	170	12	,	,	,
2163	170	13	the	the	DT
2163	170	14	best	good	JJS
2163	170	15	being	be	VBG
2163	170	16	Peng	Peng	NNP
2163	170	17	's	's	POS
2163	170	18	94	94	CD
2163	170	19	percent	percent	NN
2163	170	20	.	.	.
2163	171	1	However	however	RB
2163	171	2	,	,	,
2163	171	3	we	-PRON-	PRP
2163	171	4	achieved	achieve	VBD
2163	171	5	a	a	DT
2163	171	6	gain	gain	NN
2163	171	7	of	of	IN
2163	171	8	almost	almost	RB
2163	171	9	10	10	CD
2163	171	10	percent	percent	NN
2163	171	11	in	in	IN
2163	171	12	precision	precision	NN
2163	171	13	when	when	WRB
2163	171	14	compared	compare	VBN
2163	171	15	with	with	IN
2163	171	16	ParsCit	ParsCit	NNP
2163	171	17	(	(	-LRB-
2163	171	18	100	100	CD
2163	171	19	percent	percent	NN
2163	171	20	vs.	vs.	IN
2163	171	21	90.9	90.9	CD
2163	171	22	percent	percent	NN
2163	171	23	precision	precision	NN
2163	171	24	)	)	-RRB-
2163	171	25	.	.	.
2163	172	1	Similarly	similarly	RB
2163	172	2	,	,	,
2163	172	3	on	on	IN
2163	172	4	the	the	DT
2163	172	5	Date	Date	NNP
2163	172	6	field	field	NN
2163	172	7	,	,	,
2163	172	8	our	-PRON-	PRP$
2163	172	9	F1	f1	NN
2163	172	10	was	be	VBD
2163	172	11	98.99	98.99	CD
2163	172	12	percent	percent	NN
2163	172	13	,	,	,
2163	172	14	opposed	oppose	VBN
2163	172	15	to	to	IN
2163	172	16	ParsCit	ParsCit	NNP
2163	172	17	's	's	POS
2163	172	18	99.19	99.19	CD
2163	172	19	percent	percent	NN
2163	172	20	,	,	,
2163	172	21	but	but	CC
2163	172	22	with	with	IN
2163	172	23	a	a	DT
2163	172	24	better	well	JJR
2163	172	25	recall	recall	NN
2163	172	26	of	of	IN
2163	172	27	98.67	98.67	CD
2163	172	28	percent	percent	NN
2163	172	29	.	.	.
2163	173	1	Field	Field	NNP
2163	173	2	ParsCit	ParsCit	NNP
2163	173	3	FLUX	FLUX	NNP
2163	173	4	-	-	HYPH
2163	173	5	CIM	cim	VB
2163	173	6	Our	-PRON-	PRP$
2163	173	7	approach	approach	NN
2163	173	8	P	p	NN
2163	173	9	R	r	NN
2163	173	10	F1	f1	NN
2163	173	11	P	p	NN
2163	173	12	R	r	NN
2163	173	13	F1	f1	NN
2163	173	14	P	P	NNP
2163	173	15	R	R	NNP
2163	173	16	F1	F1	NNP
2163	173	17	Author	author	NN
2163	173	18	98.8	98.8	CD
2163	173	19	99.0	99.0	CD
2163	173	20	98.89	98.89	CD
2163	173	21	93.59	93.59	CD
2163	173	22	95.58	95.58	CD
2163	173	23	94.57	94.57	CD
2163	173	24	99.08	99.08	CD
2163	173	25	99.08	99.08	CD
2163	173	26	99.08	99.08	CD
2163	173	27	Title	Title	NNP
2163	173	28	98.8	98.8	CD
2163	173	29	98.3	98.3	CD
2163	173	30	98.54	98.54	CD
2163	173	31	93.0	93.0	CD
2163	173	32	93.0	93.0	CD
2163	173	33	93.0	93.0	CD
2163	173	34	99.65	99.65	CD
2163	173	35	99.65	99.65	CD
2163	173	36	99.65	99.65	CD
2163	173	37	Date	Date	NNP
2163	173	38	99.8	99.8	CD
2163	173	39	94.5	94.5	CD
2163	173	40	97.07	97.07	CD
2163	173	41	97.75	97.75	CD
2163	173	42	97.44	97.44	CD
2163	173	43	97.59	97.59	CD
2163	173	44	98.55	98.55	CD
2163	173	45	98.19	98.19	CD
2163	173	46	98.36	98.36	CD
2163	173	47	Pages	Pages	NNPS
2163	173	48	94.7	94.7	CD
2163	173	49	99.3	99.3	CD
2163	173	50	96.94	96.94	CD
2163	173	51	97.0	97.0	CD
2163	173	52	97.84	97.84	CD
2163	173	53	97.41	97.41	CD
2163	173	54	97.28	97.28	CD
2163	173	55	97.72	97.72	CD
2163	173	56	97.49	97.49	CD
2163	173	57	Location	Location	NNP
2163	173	58	96.9	96.9	CD
2163	173	59	88.4	88.4	CD
2163	173	60	92.45	92.45	CD
2163	173	61	96.83	96.83	CD
2163	173	62	97.6	97.6	CD
2163	173	63	97.21	97.21	CD
2163	173	64	95.55	95.55	CD
2163	173	65	94.5	94.5	CD
2163	173	66	95.02	95.02	CD
2163	173	67	Journal	Journal	NNP
2163	173	68	97.1	97.1	CD
2163	173	69	82.9	82.9	CD
2163	173	70	89.43	89.43	CD
2163	173	71	95.71	95.71	CD
2163	173	72	97.81	97.81	CD
2163	173	73	96.75	96.75	CD
2163	173	74	94.0	94.0	CD
2163	173	75	97.91	97.91	CD
2163	173	76	95.91	95.91	CD
2163	173	77	Booktitle	Booktitle	NNP
2163	173	78	95.7	95.7	CD
2163	173	79	99.3	99.3	CD
2163	173	80	97.46	97.46	CD
2163	173	81	97.47	97.47	CD
2163	173	82	95.45	95.45	CD
2163	173	83	96.45	96.45	CD
2163	173	84	99.13	99.13	CD
2163	173	85	99.13	99.13	CD
2163	173	86	99.13	99.13	CD
2163	173	87	Publisher	publisher	NN
2163	173	88	98.8	98.8	CD
2163	173	89	75.9	75.9	CD
2163	173	90	85.84	85.84	CD
2163	173	91	100	100	CD
2163	173	92	100	100	CD
2163	173	93	100	100	CD
2163	173	94	98.59	98.59	CD
2163	173	95	98.59	98.59	CD
2163	173	96	98.59	98.59	CD
2163	173	97	Table	table	NN
2163	173	98	2	2	CD
2163	173	99	.	.	.
2163	174	1	Evaluation	Evaluation	NNP
2163	174	2	Results	Results	NNP
2163	174	3	on	on	IN
2163	174	4	the	the	DT
2163	174	5	FLUX	FLUX	NNP
2163	174	6	-	-	HYPH
2163	174	7	CIM	CIM	NNP
2163	174	8	Dataset	Dataset	NNP
2163	174	9	—	—	:
2163	174	10	CS	CS	NNP
2163	174	11	Domain	Domain	NNP
2163	174	12	Field	Field	NNP
2163	174	13	FLUX	FLUX	NNP
2163	174	14	-	-	HYPH
2163	174	15	CIM	cim	VB
2163	174	16	Our	-PRON-	PRP$
2163	174	17	approach	approach	NN
2163	174	18	p	p	NN
2163	174	19	R	r	NN
2163	174	20	F1	f1	NN
2163	174	21	p	p	NN
2163	174	22	R	R	NNP
2163	174	23	F1	F1	NNP
2163	174	24	Author	author	NN
2163	174	25	98.57	98.57	CD
2163	174	26	99.04	99.04	CD
2163	174	27	98.81	98.81	CD
2163	174	28	99.8	99.8	CD
2163	174	29	99.36	99.36	CD
2163	174	30	99.57	99.57	CD
2163	174	31	Title	Title	NNP
2163	174	32	84.88	84.88	CD
2163	174	33	85.14	85.14	CD
2163	174	34	85.01	85.01	CD
2163	174	35	91.39	91.39	CD
2163	174	36	91.39	91.39	CD
2163	174	37	97.39	97.39	CD
2163	174	38	Date	Date	NNP
2163	174	39	99.85	99.85	CD
2163	174	40	99.5	99.5	CD
2163	174	41	99.61	99.61	CD
2163	174	42	99.89	99.89	CD
2163	174	43	99.69	99.69	CD
2163	174	44	99.78	99.78	CD
2163	174	45	Pages	Pages	NNPS
2163	174	46	99.1	99.1	CD
2163	174	47	99.2	99.2	CD
2163	174	48	99.45	99.45	CD
2163	174	49	99.94	99.94	CD
2163	174	50	99.59	99.59	CD
2163	174	51	99.76	99.76	CD
2163	174	52	Journal	Journal	NNP
2163	174	53	97.23	97.23	CD
2163	174	54	89.35	89.35	CD
2163	174	55	93.13	93.13	CD
2163	174	56	99.42	99.42	CD
2163	174	57	99.16	99.16	CD
2163	174	58	99.28	99.28	CD
2163	174	59	Table	Table	NNP
2163	174	60	3	3	CD
2163	174	61	.	.	.
2163	175	1	Evaluation	Evaluation	NNP
2163	175	2	Results	Results	NNP
2163	175	3	on	on	IN
2163	175	4	the	the	DT
2163	175	5	FLUX	FLUX	NNP
2163	175	6	-	-	HYPH
2163	175	7	CIM	CIM	NNP
2163	175	8	Dataset	Dataset	NNP
2163	175	9	—	—	:
2163	175	10	HS	HS	NNP
2163	175	11	Domain	Domain	NNP
2163	175	12	REFERENCE	REFERENCE	NNP
2163	175	13	INFORMATION	INFORMATION	VBD
2163	175	14	EXTRACTION	extraction	NN
2163	175	15	AND	and	CC
2163	175	16	PROCESSING	processing	NN
2163	175	17	|GROZA	|groza	NN
2163	175	18	,	,	,
2163	175	19	GRIMNES	GRIMNES	NNP
2163	175	20	,	,	,
2163	175	21	AND	and	CC
2163	175	22	HANDSCHUH	handschuh	NN
2163	175	23	16	16	CD
2163	175	24	4.1	4.1	CD
2163	175	25	Dataset	dataset	NN
2163	175	26	:	:	:
2163	175	27	FLUX	FLUX	NNP
2163	175	28	-	-	HYPH
2163	175	29	CIM	CIM	NNP
2163	175	30	FLUX	FLUX	NNP
2163	175	31	-	-	HYPH
2163	175	32	CIM	CIM	NNP
2163	175	33	43	43	CD
2163	175	34	is	be	VBZ
2163	175	35	an	an	DT
2163	175	36	unsupervised	unsupervise	VBN
2163	175	37	44	44	CD
2163	175	38	reference	reference	NN
2163	175	39	extraction	extraction	NN
2163	175	40	and	and	CC
2163	175	41	chunking	chunk	VBG
2163	175	42	system	system	NN
2163	175	43	.	.	.
2163	176	1	In	in	IN
2163	176	2	order	order	NN
2163	176	3	to	to	TO
2163	176	4	evaluate	evaluate	VB
2163	176	5	its	-PRON-	PRP$
2163	176	6	performance	performance	NN
2163	176	7	,	,	,
2163	176	8	the	the	DT
2163	176	9	authors	author	NNS
2163	176	10	of	of	IN
2163	176	11	FLUX	FLUX	NNP
2163	176	12	-	-	HYPH
2163	176	13	CIM	CIM	NNP
2163	176	14	created	create	VBD
2163	176	15	two	two	CD
2163	176	16	separate	separate	JJ
2163	176	17	datasets	dataset	NNS
2163	176	18	:	:	:
2163	176	19			:
2163	176	20	the	the	DT
2163	176	21	FLUX	FLUX	NNP
2163	176	22	-	-	HYPH
2163	176	23	CIM	CIM	NNP
2163	176	24	CS	CS	NNP
2163	176	25	dataset	dataset	NN
2163	176	26	,	,	,
2163	176	27	composed	compose	VBN
2163	176	28	on	on	IN
2163	176	29	a	a	DT
2163	176	30	collection	collection	NN
2163	176	31	of	of	IN
2163	176	32	heterogeneous	heterogeneous	JJ
2163	176	33	references	reference	NNS
2163	176	34	from	from	IN
2163	176	35	the	the	DT
2163	176	36	Computer	Computer	NNP
2163	176	37	Science	Science	NNP
2163	176	38	field	field	NN
2163	176	39	,	,	,
2163	176	40	and	and	CC
2163	176	41			:
2163	176	42	the	the	DT
2163	176	43	FLUX	FLUX	NNP
2163	176	44	-	-	HYPH
2163	176	45	CIM	CIM	NNP
2163	176	46	HS	HS	NNP
2163	176	47	dataset	dataset	NN
2163	176	48	is	be	VBZ
2163	176	49	comprised	comprise	VBN
2163	176	50	of	of	IN
2163	176	51	an	an	DT
2163	176	52	organized	organize	VBN
2163	176	53	and	and	CC
2163	176	54	controlled	controlled	JJ
2163	176	55	collection	collection	NN
2163	176	56	of	of	IN
2163	176	57	references	reference	NNS
2163	176	58	from	from	IN
2163	176	59	PubMed	PubMed	NNP
2163	176	60	.	.	.
2163	177	1	The	the	DT
2163	177	2	FLUX	FLUX	NNP
2163	177	3	-	-	HYPH
2163	177	4	CIM	CIM	NNP
2163	177	5	CS	CS	NNP
2163	177	6	dataset	dataset	NN
2163	177	7	contains	contain	VBZ
2163	177	8	three	three	CD
2163	177	9	hundred	hundred	CD
2163	177	10	reference	reference	NN
2163	177	11	strings	string	NNS
2163	177	12	randomly	randomly	RB
2163	177	13	selected	select	VBN
2163	177	14	from	from	IN
2163	177	15	the	the	DT
2163	177	16	ACM	ACM	NNP
2163	177	17	Digital	Digital	NNP
2163	177	18	Library	Library	NNP
2163	177	19	.	.	.
2163	178	1	Each	each	DT
2163	178	2	string	string	NN
2163	178	3	is	be	VBZ
2163	178	4	segmented	segment	VBN
2163	178	5	into	into	IN
2163	178	6	ten	ten	CD
2163	178	7	fields	field	NNS
2163	178	8	:	:	:
2163	178	9	Author	author	NN
2163	178	10	,	,	,
2163	178	11	Title	Title	NNP
2163	178	12	,	,	,
2163	178	13	Conf	Conf	NNP
2163	178	14	,	,	,
2163	178	15	Journal	Journal	NNP
2163	178	16	,	,	,
2163	178	17	Volume	volume	NN
2163	178	18	,	,	,
2163	178	19	Number	Number	NNP
2163	178	20	,	,	,
2163	178	21	Pub	Pub	NNP
2163	178	22	,	,	,
2163	178	23	Date	date	NN
2163	178	24	,	,	,
2163	178	25	Pages	Pages	NNPS
2163	178	26	and	and	CC
2163	178	27	Place	Place	NNP
2163	178	28	.	.	.
2163	179	1	The	the	DT
2163	179	2	FLUX	FLUX	NNP
2163	179	3	-	-	HYPH
2163	179	4	CIM	CIM	NNP
2163	179	5	HS	HS	NNP
2163	179	6	dataset	dataset	NN
2163	179	7	contains	contain	VBZ
2163	179	8	2000	2000	CD
2163	179	9	entries	entry	NNS
2163	179	10	,	,	,
2163	179	11	with	with	IN
2163	179	12	each	each	DT
2163	179	13	entry	entry	NN
2163	179	14	segmented	segment	VBD
2163	179	15	into	into	IN
2163	179	16	six	six	CD
2163	179	17	fields	field	NNS
2163	179	18	:	:	:
2163	179	19	Author	author	NN
2163	179	20	,	,	,
2163	179	21	Title	Title	NNP
2163	179	22	,	,	,
2163	179	23	Journal	Journal	NNP
2163	179	24	,	,	,
2163	179	25	Volume	volume	NN
2163	179	26	,	,	,
2163	179	27	Date	Date	NNP
2163	179	28	and	and	CC
2163	179	29	Pages	Pages	NNPS
2163	179	30	.	.	.
2163	180	1	Table	table	NN
2163	180	2	2	2	CD
2163	180	3	presents	present	VBZ
2163	180	4	the	the	DT
2163	180	5	comparative	comparative	JJ
2163	180	6	test	test	NN
2163	180	7	results	result	NNS
2163	180	8	achieved	achieve	VBN
2163	180	9	by	by	IN
2163	180	10	ParsCit	ParsCit	NNP
2163	180	11	,	,	,
2163	180	12	FLUX	FLUX	NNP
2163	180	13	-	-	HYPH
2163	180	14	CIM	CIM	NNP
2163	180	15	,	,	,
2163	180	16	and	and	CC
2163	180	17	our	-PRON-	PRP$
2163	180	18	approach	approach	NN
2163	180	19	on	on	IN
2163	180	20	the	the	DT
2163	180	21	CS	CS	NNP
2163	180	22	dataset	dataset	NN
2163	180	23	.	.	.
2163	181	1	Similar	similar	JJ
2163	181	2	to	to	IN
2163	181	3	the	the	DT
2163	181	4	CORA	CORA	NNP
2163	181	5	dataset	dataset	NN
2163	181	6	,	,	,
2163	181	7	our	-PRON-	PRP$
2163	181	8	chunker	chunker	NN
2163	181	9	outperformed	outperform	VBD
2163	181	10	the	the	DT
2163	181	11	other	other	JJ
2163	181	12	chunkers	chunker	NNS
2163	181	13	on	on	IN
2163	181	14	the	the	DT
2163	181	15	majority	majority	NN
2163	181	16	of	of	IN
2163	181	17	the	the	DT
2163	181	18	fields	field	NNS
2163	181	19	,	,	,
2163	181	20	exceptions	exception	NNS
2163	181	21	being	be	VBG
2163	181	22	the	the	DT
2163	181	23	Location	Location	NNP
2163	181	24	,	,	,
2163	181	25	Journal	Journal	NNP
2163	181	26	,	,	,
2163	181	27	and	and	CC
2163	181	28	Publisher	Publisher	NNP
2163	181	29	fields	field	NNS
2163	181	30	.	.	.
2163	182	1	The	the	DT
2163	182	2	test	test	NN
2163	182	3	results	result	VBZ
2163	182	4	on	on	IN
2163	182	5	the	the	DT
2163	182	6	HS	HS	NNP
2163	182	7	dataset	dataset	NN
2163	182	8	are	be	VBP
2163	182	9	presented	present	VBN
2163	182	10	in	in	IN
2163	182	11	table	table	NN
2163	182	12	3	3	CD
2163	182	13	.	.	.
2163	183	1	Here	here	RB
2163	183	2	we	-PRON-	PRP
2163	183	3	can	can	MD
2163	183	4	observe	observe	VB
2163	183	5	a	a	DT
2163	183	6	clear	clear	JJ
2163	183	7	performance	performance	NN
2163	183	8	improvement	improvement	NN
2163	183	9	on	on	IN
2163	183	10	all	all	DT
2163	183	11	fields	field	NNS
2163	183	12	,	,	,
2163	183	13	in	in	IN
2163	183	14	some	some	DT
2163	183	15	cases	case	NNS
2163	183	16	the	the	DT
2163	183	17	difference	difference	NN
2163	183	18	being	be	VBG
2163	183	19	significant	significant	JJ
2163	183	20	,	,	,
2163	183	21	e.g.	e.g.	RB
2163	183	22	,	,	,
2163	183	23	the	the	DT
2163	183	24	Title	title	NN
2163	183	25	field	field	NN
2163	183	26	,	,	,
2163	183	27	from	from	IN
2163	183	28	85.01	85.01	CD
2163	183	29	percent	percent	NN
2163	183	30	to	to	IN
2163	183	31	97.39	97.39	CD
2163	183	32	percent	percent	NN
2163	183	33	,	,	,
2163	183	34	or	or	CC
2163	183	35	the	the	DT
2163	183	36	Journal	Journal	NNP
2163	183	37	field	field	NN
2163	183	38	,	,	,
2163	183	39	from	from	IN
2163	183	40	93.12	93.12	CD
2163	183	41	percent	percent	NN
2163	183	42	to	to	IN
2163	183	43	99.28	99.28	CD
2163	183	44	percent	percent	NN
2163	183	45	.	.	.
2163	184	1	This	this	DT
2163	184	2	increase	increase	NN
2163	184	3	is	be	VBZ
2163	184	4	even	even	RB
2163	184	5	more	more	RBR
2163	184	6	relevant	relevant	JJ
2163	184	7	considering	consider	VBG
2163	184	8	the	the	DT
2163	184	9	size	size	NN
2163	184	10	of	of	IN
2163	184	11	the	the	DT
2163	184	12	dataset	dataset	NN
2163	184	13	,	,	,
2163	184	14	each	each	DT
2163	184	15	1percent	1percent	CD
2163	184	16	representing	represent	VBG
2163	184	17	twenty	twenty	CD
2163	184	18	references	reference	NNS
2163	184	19	.	.	.
2163	185	1	4.3	4.3	CD
2163	185	2	Dataset	dataset	NN
2163	185	3	:	:	:
2163	185	4	CS	CS	NNP
2163	185	5	-	-	HYPH
2163	185	6	SW	SW	NNP
2163	185	7	While	while	IN
2163	185	8	the	the	DT
2163	185	9	CORA	CORA	NNP
2163	185	10	and	and	CC
2163	185	11	FLUX	FLUX	NNP
2163	185	12	-	-	HYPH
2163	185	13	CIM	CIM	NNP
2163	185	14	CS	CS	NNP
2163	185	15	datasets	dataset	NNS
2163	185	16	do	do	VBP
2163	185	17	focus	focus	VB
2163	185	18	on	on	IN
2163	185	19	the	the	DT
2163	185	20	computer	computer	NN
2163	185	21	science	science	NN
2163	185	22	field	field	NN
2163	185	23	,	,	,
2163	185	24	they	-PRON-	PRP
2163	185	25	do	do	VBP
2163	185	26	not	not	RB
2163	185	27	cover	cover	VB
2163	185	28	the	the	DT
2163	185	29	slight	slight	JJ
2163	185	30	differences	difference	NNS
2163	185	31	in	in	IN
2163	185	32	reference	reference	NN
2163	185	33	format	format	NN
2163	185	34	that	that	WDT
2163	185	35	can	can	MD
2163	185	36	be	be	VB
2163	185	37	found	find	VBN
2163	185	38	nowadays	nowadays	RB
2163	185	39	in	in	IN
2163	185	40	the	the	DT
2163	185	41	Semantic	semantic	JJ
2163	185	42	Web	web	NN
2163	185	43	community	community	NN
2163	185	44	.	.	.
2163	186	1	Consequently	consequently	RB
2163	186	2	,	,	,
2163	186	3	to	to	TO
2163	186	4	show	show	VB
2163	186	5	the	the	DT
2163	186	6	even	even	RB
2163	186	7	broader	broad	JJR
2163	186	8	application	application	NN
2163	186	9	of	of	IN
2163	186	10	our	-PRON-	PRP$
2163	186	11	approach	approach	NN
2163	186	12	,	,	,
2163	186	13	we	-PRON-	PRP
2163	186	14	have	have	VBP
2163	186	15	compiled	compile	VBN
2163	186	16	a	a	DT
2163	186	17	dataset	dataset	NN
2163	186	18	named	name	VBN
2163	186	19	CS	CS	NNP
2163	186	20	-	-	HYPH
2163	186	21	SW	SW	NNP
2163	186	22	comprising	comprise	VBG
2163	186	23	576	576	CD
2163	186	24	reference	reference	NN
2163	186	25	strings	string	NNS
2163	186	26	randomly	randomly	RB
2163	186	27	selected	select	VBN
2163	186	28	from	from	IN
2163	186	29	publications	publication	NNS
2163	186	30	in	in	IN
2163	186	31	the	the	DT
2163	186	32	Semantic	Semantic	NNP
2163	186	33	Web	Web	NNP
2163	186	34	area	area	NN
2163	186	35	,	,	,
2163	186	36	from	from	IN
2163	186	37	conferences	conference	NNS
2163	186	38	such	such	JJ
2163	186	39	as	as	IN
2163	186	40	International	International	NNP
2163	186	41	Semantic	Semantic	NNP
2163	186	42	Web	Web	NNP
2163	186	43	Conference	Conference	NNP
2163	186	44	(	(	-LRB-
2163	186	45	ISWC	ISWC	NNP
2163	186	46	)	)	-RRB-
2163	186	47	,	,	,
2163	186	48	the	the	DT
2163	186	49	European	European	NNP
2163	186	50	Semantic	Semantic	NNP
2163	186	51	Web	Web	NNP
2163	186	52	Conference	Conference	NNP
2163	186	53	(	(	-LRB-
2163	186	54	ESWC	ESWC	NNP
2163	186	55	)	)	-RRB-
2163	186	56	,	,	,
2163	186	57	the	the	DT
2163	186	58	World	World	NNP
2163	186	59	Wide	Wide	NNP
2163	186	60	Web	Web	NNP
2163	186	61	Conference	Conference	NNP
2163	186	62	(	(	-LRB-
2163	186	63	WWW	WWW	NNP
2163	186	64	)	)	-RRB-
2163	186	65	,	,	,
2163	186	66	or	or	CC
2163	186	67	the	the	DT
2163	186	68	European	European	NNP
2163	186	69	Conference	Conference	NNP
2163	186	70	on	on	IN
2163	186	71	Knowledge	Knowledge	NNP
2163	186	72	Acquisition	acquisition	NN
2163	186	73	(	(	-LRB-
2163	186	74	and	and	CC
2163	186	75	co	co	JJ
2163	186	76	-	-	JJ
2163	186	77	located	locate	VBN
2163	186	78	workshops	workshop	NNS
2163	186	79	)	)	-RRB-
2163	186	80	.	.	.
2163	187	1	45	45	CD
2163	187	2	Each	each	DT
2163	187	3	reference	reference	NN
2163	187	4	entry	entry	NN
2163	187	5	is	be	VBZ
2163	187	6	segmented	segment	VBN
2163	187	7	into	into	IN
2163	187	8	twelve	twelve	CD
2163	187	9	fields	field	NNS
2163	187	10	:	:	:
2163	187	11	Author	author	NN
2163	187	12	,	,	,
2163	187	13	Title	Title	NNP
2163	187	14	,	,	,
2163	187	15	Conference	Conference	NNP
2163	187	16	,	,	,
2163	187	17	Workshop	Workshop	NNP
2163	187	18	,	,	,
2163	187	19	Journal	Journal	NNP
2163	187	20	,	,	,
2163	187	21	Techrep	Techrep	NNP
2163	187	22	,	,	,
2163	187	23	Organization	Organization	NNP
2163	187	24	,	,	,
2163	187	25	Publisher	Publisher	NNP
2163	187	26	,	,	,
2163	187	27	Date	date	NN
2163	187	28	,	,	,
2163	187	29	Pages	Pages	NNPS
2163	187	30	,	,	,
2163	187	31	Website	Website	NNP
2163	187	32	and	and	CC
2163	187	33	Location	Location	NNP
2163	187	34	.	.	.
2163	188	1	Table	table	NN
2163	188	2	4	4	CD
2163	188	3	shows	show	VBZ
2163	188	4	the	the	DT
2163	188	5	results	result	NNS
2163	188	6	of	of	IN
2163	188	7	the	the	DT
2163	188	8	tests	test	NNS
2163	188	9	carried	carry	VBN
2163	188	10	out	out	RP
2163	188	11	on	on	IN
2163	188	12	this	this	DT
2163	188	13	dataset	dataset	NN
2163	188	14	.	.	.
2163	189	1	One	one	PRP
2163	189	2	can	can	MD
2163	189	3	easily	easily	RB
2163	189	4	observe	observe	VB
2163	189	5	that	that	IN
2163	189	6	the	the	DT
2163	189	7	chunker	chunker	NN
2163	189	8	performed	perform	VBN
2163	189	9	in	in	IN
2163	189	10	a	a	DT
2163	189	11	similar	similar	JJ
2163	189	12	manner	manner	NN
2163	189	13	as	as	IN
2163	189	14	on	on	IN
2163	189	15	the	the	DT
2163	189	16	CORA	CORA	NNP
2163	189	17	dataset	dataset	NN
2163	189	18	,	,	,
2163	189	19	with	with	IN
2163	189	20	emphasis	emphasis	NN
2163	189	21	on	on	IN
2163	189	22	the	the	DT
2163	189	23	Author	author	NN
2163	189	24	,	,	,
2163	189	25	Date	Date	NNP
2163	189	26	,	,	,
2163	189	27	Pages	Pages	NNPS
2163	189	28	and	and	CC
2163	189	29	Publisher	Publisher	NNP
2163	189	30	fields	field	NNS
2163	189	31	.	.	.
2163	190	1	Field	field	VB
2163	190	2	Our	-PRON-	PRP$
2163	190	3	approach	approach	NN
2163	190	4	P	P	NNP
2163	190	5	R	R	NNP
2163	190	6	F1	F1	NNP
2163	190	7	Author	author	NN
2163	190	8	98.61	98.61	CD
2163	190	9	99.27	99.27	CD
2163	190	10	98.93	98.93	CD
2163	190	11	Title	Title	NNP
2163	190	12	94.91	94.91	CD
2163	190	13	93.29	93.29	CD
2163	190	14	94.09	94.09	CD
2163	190	15	Date	date	NN
2163	190	16	98.89	98.89	CD
2163	190	17	98.34	98.34	CD
2163	190	18	98.61	98.61	CD
2163	190	19	Pages	Pages	NNPS
2163	190	20	98.94	98.94	CD
2163	190	21	97.24	97.24	CD
2163	190	22	98.08	98.08	CD
2163	190	23	Location	Location	NNP
2163	190	24	93.9	93.9	CD
2163	190	25	92.77	92.77	CD
2163	190	26	93.33	93.33	CD
2163	190	27	Organization	Organization	NNP
2163	190	28	85.71	85.71	CD
2163	190	29	80	80	CD
2163	190	30	00	00	CD
2163	190	31	82.75	82.75	CD
2163	190	32	Journal	Journal	NNP
2163	190	33	94.59	94.59	CD
2163	190	34	93.33	93.33	CD
2163	190	35	93.95	93.95	CD
2163	190	36	INFORMATION	INFORMATION	NNP
2163	190	37	TECHNOLOGY	technology	NN
2163	190	38	AND	and	CC
2163	190	39	LIBRARIES	LIBRARIES	NNP
2163	190	40	|	|	NNP
2163	190	41	JUNE	JUNE	NNP
2163	190	42	2012	2012	CD
2163	190	43	17	17	CD
2163	190	44	Conference	Conference	NNP
2163	190	45	96.66	96.66	CD
2163	190	46	95.08	95.08	CD
2163	190	47	95.86	95.86	CD
2163	190	48	Workshop	Workshop	NNP
2163	190	49	83.33	83.33	CD
2163	190	50	88.23	88.23	CD
2163	190	51	85.71	85.71	CD
2163	190	52	Publisher	Publisher	NNP
2163	190	53	96.61	96.61	CD
2163	190	54	97.43	97.43	CD
2163	190	55	97.01	97.01	CD
2163	190	56	Tech	Tech	NNP
2163	190	57	.	.	.
2163	191	1	rep	rep	NNP
2163	191	2	.	.	NNP
2163	191	3	100	100	CD
2163	191	4	80	80	CD
2163	191	5	88.88	88.88	CD
2163	191	6	Website	Website	NNP
2163	191	7	98.14	98.14	CD
2163	191	8	94.64	94.64	CD
2163	191	9	96.35	96.35	CD
2163	191	10	Table	Table	NNP
2163	191	11	4	4	CD
2163	191	12	.	.	.
2163	192	1	Evaluation	Evaluation	NNP
2163	192	2	Results	Results	NNP
2163	192	3	on	on	IN
2163	192	4	the	the	DT
2163	192	5	CS	CS	NNP
2163	192	6	-	-	HYPH
2163	192	7	SW	SW	NNP
2163	192	8	Dataset	Dataset	NNP
2163	192	9	5	5	CD
2163	192	10	.	.	.
2163	193	1	CONCLUSION	CONCLUSION	NNP
2163	193	2	In	in	IN
2163	193	3	this	this	DT
2163	193	4	paper	paper	NN
2163	193	5	we	-PRON-	PRP
2163	193	6	presented	present	VBD
2163	193	7	a	a	DT
2163	193	8	novel	novel	JJ
2163	193	9	approach	approach	NN
2163	193	10	for	for	IN
2163	193	11	extracting	extract	VBG
2163	193	12	and	and	CC
2163	193	13	chunking	chunk	VBG
2163	193	14	reference	reference	NN
2163	193	15	information	information	NN
2163	193	16	from	from	IN
2163	193	17	scientific	scientific	JJ
2163	193	18	publications	publication	NNS
2163	193	19	.	.	.
2163	194	1	The	the	DT
2163	194	2	solution	solution	NN
2163	194	3	,	,	,
2163	194	4	realized	realize	VBD
2163	194	5	using	use	VBG
2163	194	6	a	a	DT
2163	194	7	CRF	CRF	NNP
2163	194	8	trained	train	VBN
2163	194	9	chunker	chunker	NN
2163	194	10	,	,	,
2163	194	11	achieved	achieve	VBD
2163	194	12	good	good	JJ
2163	194	13	results	result	NNS
2163	194	14	in	in	IN
2163	194	15	the	the	DT
2163	194	16	experimental	experimental	JJ
2163	194	17	evaluation	evaluation	NN
2163	194	18	,	,	,
2163	194	19	in	in	IN
2163	194	20	addition	addition	NN
2163	194	21	to	to	IN
2163	194	22	an	an	DT
2163	194	23	increased	increase	VBN
2163	194	24	versatility	versatility	NN
2163	194	25	shown	show	VBN
2163	194	26	by	by	IN
2163	194	27	applying	apply	VBG
2163	194	28	the	the	DT
2163	194	29	one	one	CD
2163	194	30	-	-	HYPH
2163	194	31	time	time	NN
2163	194	32	trained	train	VBN
2163	194	33	chunker	chunker	NN
2163	194	34	on	on	IN
2163	194	35	multiple	multiple	JJ
2163	194	36	testing	testing	NN
2163	194	37	datasets	dataset	NNS
2163	194	38	.	.	.
2163	195	1	This	this	DT
2163	195	2	enables	enable	VBZ
2163	195	3	a	a	DT
2163	195	4	straightforward	straightforward	JJ
2163	195	5	adoption	adoption	NN
2163	195	6	and	and	CC
2163	195	7	reuse	reuse	NN
2163	195	8	of	of	IN
2163	195	9	our	-PRON-	PRP$
2163	195	10	solution	solution	NN
2163	195	11	for	for	IN
2163	195	12	generating	generate	VBG
2163	195	13	semantic	semantic	JJ
2163	195	14	metadata	metadata	NN
2163	195	15	in	in	IN
2163	195	16	any	any	DT
2163	195	17	digital	digital	JJ
2163	195	18	library	library	NN
2163	195	19	or	or	CC
2163	195	20	publication	publication	NN
2163	195	21	repository	repository	NN
2163	195	22	focused	focus	VBD
2163	195	23	on	on	IN
2163	195	24	scientific	scientific	JJ
2163	195	25	publishing	publishing	NN
2163	195	26	.	.	.
2163	196	1	As	as	IN
2163	196	2	next	next	JJ
2163	196	3	steps	step	NNS
2163	196	4	,	,	,
2163	196	5	we	-PRON-	PRP
2163	196	6	plan	plan	VBP
2163	196	7	to	to	TO
2163	196	8	create	create	VB
2163	196	9	a	a	DT
2163	196	10	comprehensive	comprehensive	JJ
2163	196	11	dataset	dataset	NN
2163	196	12	covering	cover	VBG
2163	196	13	multiple	multiple	JJ
2163	196	14	heterogeneous	heterogeneous	JJ
2163	196	15	domains	domain	NNS
2163	196	16	(	(	-LRB-
2163	196	17	e.g.	e.g.	RB
2163	196	18	,	,	,
2163	196	19	social	social	JJ
2163	196	20	sciences	sciences	NNPS
2163	196	21	or	or	CC
2163	196	22	digital	digital	JJ
2163	196	23	humanities	humanity	NNS
2163	196	24	)	)	-RRB-
2163	196	25	and	and	CC
2163	196	26	evaluate	evaluate	VB
2163	196	27	the	the	DT
2163	196	28	chunker	chunker	NN
2163	196	29	’s	’s	POS
2163	196	30	performance	performance	NN
2163	196	31	on	on	IN
2163	196	32	it	-PRON-	PRP
2163	196	33	.	.	.
2163	197	1	Then	then	RB
2163	197	2	we	-PRON-	PRP
2163	197	3	will	will	MD
2163	197	4	focus	focus	VB
2163	197	5	on	on	IN
2163	197	6	developing	develop	VBG
2163	197	7	an	an	DT
2163	197	8	accurate	accurate	JJ
2163	197	9	reference	reference	NN
2163	197	10	consolidation	consolidation	NN
2163	197	11	and	and	CC
2163	197	12	linking	linking	NN
2163	197	13	technique	technique	NN
2163	197	14	,	,	,
2163	197	15	to	to	TO
2163	197	16	address	address	VB
2163	197	17	the	the	DT
2163	197	18	second	second	JJ
2163	197	19	step	step	NN
2163	197	20	mentioned	mention	VBN
2163	197	21	in	in	IN
2163	197	22	section	section	NN
2163	197	23	1	1	CD
2163	197	24	,	,	,
2163	197	25	i.e.	i.e.	FW
2163	197	26	,	,	,
2163	197	27	aligning	align	VBG
2163	197	28	the	the	DT
2163	197	29	resulting	result	VBG
2163	197	30	metadata	metadata	NN
2163	197	31	to	to	IN
2163	197	32	the	the	DT
2163	197	33	existing	exist	VBG
2163	197	34	Linked	Linked	NNP
2163	197	35	Data	Data	NNP
2163	197	36	on	on	IN
2163	197	37	the	the	DT
2163	197	38	Web	web	NN
2163	197	39	.	.	.
2163	198	1	We	-PRON-	PRP
2163	198	2	plan	plan	VBP
2163	198	3	to	to	TO
2163	198	4	develop	develop	VB
2163	198	5	a	a	DT
2163	198	6	flexible	flexible	JJ
2163	198	7	consolidation	consolidation	NN
2163	198	8	mechanism	mechanism	NN
2163	198	9	by	by	IN
2163	198	10	dynamically	dynamically	RB
2163	198	11	generating	generate	VBG
2163	198	12	and	and	CC
2163	198	13	executing	execute	VBG
2163	198	14	SPARQL	sparql	JJ
2163	198	15	queries	query	NNS
2163	198	16	from	from	IN
2163	198	17	chunked	chunk	VBN
2163	198	18	reference	reference	NN
2163	198	19	fields	field	NNS
2163	198	20	and	and	CC
2163	198	21	filtering	filter	VBG
2163	198	22	the	the	DT
2163	198	23	results	result	NNS
2163	198	24	via	via	IN
2163	198	25	two	two	CD
2163	198	26	string	string	NN
2163	198	27	approximation	approximation	NN
2163	198	28	metrics	metric	NNS
2163	198	29	(	(	-LRB-
2163	198	30	a	a	DT
2163	198	31	combination	combination	NN
2163	198	32	of	of	IN
2163	198	33	Monge	Monge	NNP
2163	198	34	-	-	HYPH
2163	198	35	Elkan	Elkan	NNP
2163	198	36	and	and	CC
2163	198	37	Chapman	Chapman	NNP
2163	198	38	Soundex	Soundex	NNP
2163	198	39	algorithms	algorithm	NNS
2163	198	40	)	)	-RRB-
2163	198	41	.	.	.
2163	199	1	The	the	DT
2163	199	2	SPARQL	sparql	NN
2163	199	3	queries	query	NNS
2163	199	4	generation	generation	NN
2163	199	5	will	will	MD
2163	199	6	be	be	VB
2163	199	7	implemented	implement	VBN
2163	199	8	in	in	IN
2163	199	9	an	an	DT
2163	199	10	extensible	extensible	JJ
2163	199	11	manner	manner	NN
2163	199	12	,	,	,
2163	199	13	via	via	IN
2163	199	14	customizable	customizable	JJ
2163	199	15	query	query	NN
2163	199	16	modules	module	NNS
2163	199	17	,	,	,
2163	199	18	to	to	TO
2163	199	19	accommodate	accommodate	VB
2163	199	20	the	the	DT
2163	199	21	heterogeneous	heterogeneous	JJ
2163	199	22	nature	nature	NN
2163	199	23	of	of	IN
2163	199	24	the	the	DT
2163	199	25	diverse	diverse	JJ
2163	199	26	Linked	Linked	NNP
2163	199	27	Data	Data	NNP
2163	199	28	sources	source	NNS
2163	199	29	.	.	.
2163	200	1	Finally	finally	RB
2163	200	2	,	,	,
2163	200	3	we	-PRON-	PRP
2163	200	4	intend	intend	VBP
2163	200	5	to	to	TO
2163	200	6	develop	develop	VB
2163	200	7	an	an	DT
2163	200	8	overlay	overlay	NN
2163	200	9	interface	interface	NN
2163	200	10	for	for	IN
2163	200	11	arbitrary	arbitrary	JJ
2163	200	12	online	online	JJ
2163	200	13	publication	publication	NN
2163	200	14	repositories	repository	NNS
2163	200	15	,	,	,
2163	200	16	to	to	TO
2163	200	17	enable	enable	VB
2163	200	18	on	on	IN
2163	200	19	-	-	HYPH
2163	200	20	the	the	DT
2163	200	21	-	-	HYPH
2163	200	22	fly	fly	NN
2163	200	23	creation	creation	NN
2163	200	24	,	,	,
2163	200	25	visualization	visualization	NN
2163	200	26	,	,	,
2163	200	27	and	and	CC
2163	200	28	linking	linking	NN
2163	200	29	of	of	IN
2163	200	30	semantic	semantic	JJ
2163	200	31	metadata	metadata	NN
2163	200	32	from	from	IN
2163	200	33	repositories	repository	NNS
2163	200	34	that	that	WDT
2163	200	35	currently	currently	RB
2163	200	36	do	do	VBP
2163	200	37	not	not	RB
2163	200	38	expose	expose	VB
2163	200	39	their	-PRON-	PRP$
2163	200	40	datasets	dataset	NNS
2163	200	41	in	in	IN
2163	200	42	a	a	DT
2163	200	43	semantic	semantic	JJ
2163	200	44	/	/	SYM
2163	200	45	linked	link	VBN
2163	200	46	manner	manner	NN
2163	200	47	.	.	.
2163	201	1	ACKNOWLEDGEMENTS	ACKNOWLEDGEMENTS	NNP
2163	201	2	The	the	DT
2163	201	3	work	work	NN
2163	201	4	presented	present	VBN
2163	201	5	in	in	IN
2163	201	6	this	this	DT
2163	201	7	paper	paper	NN
2163	201	8	has	have	VBZ
2163	201	9	been	be	VBN
2163	201	10	funded	fund	VBN
2163	201	11	by	by	IN
2163	201	12	Science	Science	NNP
2163	201	13	Foundation	Foundation	NNP
2163	201	14	Ireland	Ireland	NNP
2163	201	15	under	under	IN
2163	201	16	Grant	Grant	NNP
2163	201	17	No	No	NNP
2163	201	18	.	.	.
2163	202	1	SFI/08	sfi/08	JJ
2163	202	2	/	/	SYM
2163	202	3	CE	CE	NNP
2163	202	4	/	/	SYM
2163	202	5	I1380	I1380	NNP
2163	202	6	(	(	-LRB-
2163	202	7	Lion-2	Lion-2	NNP
2163	202	8	)	)	-RRB-
2163	202	9	.	.	.
2163	203	1	REFERENCES	reference	NNS
2163	203	2	AND	and	CC
2163	203	3	NOTES	note	NNS
2163	203	4	1	1	CD
2163	203	5	.	.	.
2163	204	1	Tim	Tim	NNP
2163	204	2	Berners	Berners	NNP
2163	204	3	-	-	HYPH
2163	204	4	Lee	Lee	NNP
2163	204	5	et	et	NNP
2163	204	6	al	al	NNP
2163	204	7	.	.	NNP
2163	204	8	,	,	,
2163	204	9	“	"	``
2163	204	10	The	the	DT
2163	204	11	Semantic	semantic	JJ
2163	204	12	Web	web	NN
2163	204	13	,	,	,
2163	204	14	”	"	''
2163	204	15	Scientific	scientific	JJ
2163	204	16	American	American	NNP
2163	204	17	284	284	CD
2163	204	18	(	(	-LRB-
2163	204	19	2001	2001	CD
2163	204	20	)	)	-RRB-
2163	204	21	:	:	:
2163	204	22	35–43	35–43	NN
2163	204	23	.	.	.
2163	205	1	2	2	LS
2163	205	2	.	.	.
2163	206	1	Christian	Christian	NNP
2163	206	2	Bizer	Bizer	NNP
2163	206	3	et	et	NNP
2163	206	4	al	al	NNP
2163	206	5	.	.	NNP
2163	206	6	,	,	,
2163	206	7	“	"	``
2163	206	8	Linked	link	VBN
2163	206	9	Data	Data	NNP
2163	206	10	—	—	:
2163	206	11	The	The	NNP
2163	206	12	Story	Story	NNP
2163	206	13	So	so	RB
2163	206	14	Far	far	RB
2163	206	15	,	,	,
2163	206	16	”	"	''
2163	206	17	International	International	NNP
2163	206	18	Journal	Journal	NNP
2163	206	19	on	on	IN
2163	206	20	Semantic	Semantic	NNP
2163	206	21	Web	Web	NNP
2163	206	22	and	and	CC
2163	206	23	Information	Information	NNP
2163	206	24	Systems	Systems	NNP
2163	206	25	5	5	CD
2163	206	26	(	(	-LRB-
2163	206	27	2009	2009	CD
2163	206	28	)	)	-RRB-
2163	206	29	:	:	:
2163	206	30	1–22	1–22	CD
2163	206	31	.	.	.
2163	207	1	3	3	LS
2163	207	2	.	.	.
2163	208	1	Generating	generate	VBG
2163	208	2	computer	computer	NN
2163	208	3	-	-	HYPH
2163	208	4	understandable	understandable	JJ
2163	208	5	metadata	metadata	NN
2163	208	6	represents	represent	VBZ
2163	208	7	an	an	DT
2163	208	8	issue	issue	NN
2163	208	9	,	,	,
2163	208	10	in	in	IN
2163	208	11	general	general	JJ
2163	208	12	,	,	,
2163	208	13	in	in	IN
2163	208	14	the	the	DT
2163	208	15	publishing	publishing	NN
2163	208	16	domain	domain	NN
2163	208	17	,	,	,
2163	208	18	and	and	CC
2163	208	19	not	not	RB
2163	208	20	necessarily	necessarily	RB
2163	208	21	only	only	RB
2163	208	22	in	in	IN
2163	208	23	its	-PRON-	PRP$
2163	208	24	scientific	scientific	JJ
2163	208	25	area	area	NN
2163	208	26	.	.	.
2163	209	1	However	however	RB
2163	209	2	,	,	,
2163	209	3	the	the	DT
2163	209	4	relevant	relevant	JJ
2163	209	5	literature	literature	NN
2163	209	6	dealing	deal	VBG
2163	209	7	with	with	IN
2163	209	8	metadata	metadata	NN
2163	209	9	extraction	extraction	NN
2163	209	10	/	/	SYM
2163	209	11	generation	generation	NN
2163	209	12	has	have	VBZ
2163	209	13	focused	focus	VBN
2163	209	14	on	on	IN
2163	209	15	scientific	scientific	JJ
2163	209	16	publishing	publishing	NN
2163	209	17	,	,	,
2163	209	18	because	because	IN
2163	209	19	of	of	IN
2163	209	20	its	-PRON-	PRP$
2163	209	21	accelerated	accelerated	JJ
2163	209	22	growing	grow	VBG
2163	209	23	rate	rate	NN
2163	209	24	,	,	,
2163	209	25	especially	especially	RB
2163	209	26	with	with	IN
2163	209	27	the	the	DT
2163	209	28	increasing	increase	VBG
2163	209	29	use	use	NN
2163	209	30	of	of	IN
2163	209	31	the	the	DT
2163	209	32	World	World	NNP
2163	209	33	Wide	Wide	NNP
2163	209	34	Web	Web	NNP
2163	209	35	as	as	IN
2163	209	36	a	a	DT
2163	209	37	dissemination	dissemination	NN
2163	209	38	mechanism	mechanism	NN
2163	209	39	.	.	.
2163	210	1	REFERENCE	reference	NN
2163	210	2	INFORMATION	INFORMATION	NNS
2163	210	3	EXTRACTION	extraction	NN
2163	210	4	AND	and	CC
2163	210	5	PROCESSING	processing	NN
2163	210	6	|GROZA	|groza	NN
2163	210	7	,	,	,
2163	210	8	GRIMNES	GRIMNES	NNP
2163	210	9	,	,	,
2163	210	10	AND	and	CC
2163	210	11	HANDSCHUH	HANDSCHUH	NNP
2163	210	12	18	18	CD
2163	210	13	4	4	CD
2163	210	14	.	.	.
2163	211	1	Knud	Knud	NNP
2163	211	2	Moeller	Moeller	NNP
2163	211	3	et	et	FW
2163	211	4	al	al	NNP
2163	211	5	.	.	NNP
2163	211	6	,	,	,
2163	211	7	“	"	``
2163	211	8	Recipes	recipe	NNS
2163	211	9	for	for	IN
2163	211	10	Semantic	Semantic	NNP
2163	211	11	Web	Web	NNP
2163	211	12	Dog	Dog	NNP
2163	211	13	Food	Food	NNP
2163	211	14	–	–	:
2163	211	15	The	the	DT
2163	211	16	ESWC	eswc	NN
2163	211	17	and	and	CC
2163	211	18	ISWC	ISWC	NNP
2163	211	19	Metadata	Metadata	NNP
2163	211	20	Projects	Projects	NNPS
2163	211	21	,	,	,
2163	211	22	”	"	''
2163	211	23	Proceedings	proceeding	NNS
2163	211	24	of	of	IN
2163	211	25	the	the	DT
2163	211	26	6th	6th	JJ
2163	211	27	International	International	NNP
2163	211	28	Semantic	Semantic	NNP
2163	211	29	Web	Web	NNP
2163	211	30	Conference	Conference	NNP
2163	211	31	(	(	-LRB-
2163	211	32	Busan	Busan	NNP
2163	211	33	,	,	,
2163	211	34	Korea	Korea	NNP
2163	211	35	,	,	,
2163	211	36	2007	2007	CD
2163	211	37	)	)	-RRB-
2163	211	38	.	.	.
2163	212	1	5	5	CD
2163	212	2	.	.	.
2163	213	1	Wei	Wei	NNP
2163	213	2	Peng	Peng	NNP
2163	213	3	and	and	CC
2163	213	4	Tao	Tao	NNP
2163	213	5	Li	Li	NNP
2163	213	6	,	,	,
2163	213	7	“	"	``
2163	213	8	Temporal	temporal	JJ
2163	213	9	relation	relation	NN
2163	213	10	co	co	JJ
2163	213	11	-	-	JJ
2163	213	12	clustering	clustering	NN
2163	213	13	on	on	IN
2163	213	14	directional	directional	JJ
2163	213	15	social	social	JJ
2163	213	16	network	network	NN
2163	213	17	and	and	CC
2163	213	18	author	author	NN
2163	213	19	-	-	HYPH
2163	213	20	topic	topic	NN
2163	213	21	evolution	evolution	NN
2163	213	22	,	,	,
2163	213	23	”	"	''
2163	213	24	Knowledge	Knowledge	NNP
2163	213	25	and	and	CC
2163	213	26	Information	Information	NNP
2163	213	27	Systems	Systems	NNP
2163	213	28	26	26	CD
2163	213	29	(	(	-LRB-
2163	213	30	2011	2011	CD
2163	213	31	)	)	-RRB-
2163	213	32	:	:	:
2163	213	33	467–86	467–86	CD
2163	213	34	.	.	.
2163	214	1	6	6	CD
2163	214	2	.	.	.
2163	215	1	Laszlo	Laszlo	NNP
2163	215	2	Barabasi	Barabasi	NNP
2163	215	3	et	et	FW
2163	215	4	al	al	NNP
2163	215	5	.	.	NNP
2163	215	6	,	,	,
2163	215	7	“	"	``
2163	215	8	Evolution	evolution	NN
2163	215	9	of	of	IN
2163	215	10	the	the	DT
2163	215	11	social	social	JJ
2163	215	12	network	network	NN
2163	215	13	of	of	IN
2163	215	14	scientific	scientific	JJ
2163	215	15	collaborations	collaboration	NNS
2163	215	16	,	,	,
2163	215	17	”	"	''
2163	215	18	Physica	Physica	NNP
2163	215	19	A	A	NNP
2163	215	20	:	:	:
2163	215	21	Statistical	Statistical	NNP
2163	215	22	Mechanics	Mechanics	NNPS
2163	215	23	and	and	CC
2163	215	24	its	-PRON-	PRP$
2163	215	25	Applications	Applications	NNPS
2163	215	26	311	311	CD
2163	215	27	(	(	-LRB-
2163	215	28	2002	2002	CD
2163	215	29	)	)	-RRB-
2163	215	30	:	:	:
2163	215	31	590–614	590–614	CD
2163	215	32	.	.	.
2163	216	1	7	7	LS
2163	216	2	.	.	.
2163	217	1	Xiaoming	Xiaoming	NNP
2163	217	2	Liu	Liu	NNP
2163	217	3	et	et	NNP
2163	217	4	al	al	NNP
2163	217	5	.	.	NNP
2163	217	6	,	,	,
2163	217	7	“	"	``
2163	217	8	Co	co	NN
2163	217	9	-	-	NN
2163	217	10	authorship	authorship	NN
2163	217	11	networks	network	NNS
2163	217	12	in	in	IN
2163	217	13	the	the	DT
2163	217	14	digital	digital	JJ
2163	217	15	library	library	NN
2163	217	16	research	research	NN
2163	217	17	community	community	NN
2163	217	18	,	,	,
2163	217	19	”	"	''
2163	217	20	Information	Information	NNP
2163	217	21	Processing	Processing	NNP
2163	217	22	&	&	CC
2163	217	23	Management	Management	NNP
2163	217	24	41	41	CD
2163	217	25	(	(	-LRB-
2163	217	26	2005	2005	CD
2163	217	27	)	)	-RRB-
2163	217	28	:	:	:
2163	217	29	1462–80	1462–80	LS
2163	217	30	.	.	.
2163	218	1	8	8	LS
2163	218	2	.	.	.
2163	219	1	John	John	NNP
2163	219	2	D.	D.	NNP
2163	219	3	Lafferty	Lafferty	NNP
2163	219	4	et	et	NNP
2163	219	5	al	al	NNP
2163	219	6	.	.	NNP
2163	219	7	,	,	,
2163	219	8	“	"	``
2163	219	9	Conditional	Conditional	NNP
2163	219	10	Random	Random	NNP
2163	219	11	Fields	Fields	NNPS
2163	219	12	:	:	:
2163	219	13	Probabilistic	Probabilistic	NNP
2163	219	14	Models	Models	NNPS
2163	219	15	for	for	IN
2163	219	16	Segmenting	segmenting	NN
2163	219	17	and	and	CC
2163	219	18	Labeling	Labeling	NNP
2163	219	19	Sequence	Sequence	NNP
2163	219	20	Data	Data	NNP
2163	219	21	,	,	,
2163	219	22	”	"	''
2163	219	23	Proceedings	proceeding	NNS
2163	219	24	of	of	IN
2163	219	25	the	the	DT
2163	219	26	18th	18th	JJ
2163	219	27	International	International	NNP
2163	219	28	Conference	Conference	NNP
2163	219	29	on	on	IN
2163	219	30	Machine	Machine	NNP
2163	219	31	Learning	Learning	NNP
2163	219	32	(	(	-LRB-
2163	219	33	San	San	NNP
2163	219	34	Francisco	Francisco	NNP
2163	219	35	,	,	,
2163	219	36	CA	CA	NNP
2163	219	37	,	,	,
2163	219	38	USA	USA	NNP
2163	219	39	,	,	,
2163	219	40	2001	2001	CD
2163	219	41	)	)	-RRB-
2163	219	42	:	:	:
2163	219	43	282–89	282–89	CD
2163	219	44	.	.	.
2163	220	1	9	9	CD
2163	220	2	.	.	.
2163	221	1	Vladimir	Vladimir	NNP
2163	221	2	Vapnik	Vapnik	NNP
2163	221	3	,	,	,
2163	221	4	The	the	DT
2163	221	5	Nature	Nature	NNP
2163	221	6	of	of	IN
2163	221	7	Statistical	statistical	JJ
2163	221	8	Learning	Learning	NNP
2163	221	9	Theory	Theory	NNP
2163	221	10	(	(	-LRB-
2163	221	11	New	New	NNP
2163	221	12	York	York	NNP
2163	221	13	:	:	:
2163	221	14	Springer	Springer	NNP
2163	221	15	,	,	,
2163	221	16	1995	1995	CD
2163	221	17	)	)	-RRB-
2163	221	18	.	.	.
2163	222	1	10	10	CD
2163	222	2	.	.	.
2163	223	1	Isaac	Isaac	NNP
2163	223	2	G.	G.	NNP
2163	223	3	Councill	Councill	NNP
2163	223	4	et	et	FW
2163	223	5	al	al	NNP
2163	223	6	.	.	NNP
2163	223	7	,	,	,
2163	223	8	“	"	``
2163	223	9	ParsCit	parscit	NN
2163	223	10	:	:	:
2163	223	11	An	an	DT
2163	223	12	Open	open	JJ
2163	223	13	-	-	HYPH
2163	223	14	source	source	NN
2163	223	15	CRF	CRF	NNP
2163	223	16	Reference	Reference	NNP
2163	223	17	String	String	NNP
2163	223	18	Parsing	Parsing	NNP
2163	223	19	Package	package	NN
2163	223	20	,	,	,
2163	223	21	”	"	''
2163	223	22	Proceedings	proceeding	NNS
2163	223	23	of	of	IN
2163	223	24	the	the	DT
2163	223	25	Sixth	Sixth	NNP
2163	223	26	International	International	NNP
2163	223	27	Language	Language	NNP
2163	223	28	Resources	Resources	NNPS
2163	223	29	and	and	CC
2163	223	30	Evaluation	Evaluation	NNP
2163	223	31	(	(	-LRB-
2163	223	32	Marrakech	Marrakech	NNP
2163	223	33	,	,	,
2163	223	34	Morocco	Morocco	NNP
2163	223	35	,	,	,
2163	223	36	2008	2008	CD
2163	223	37	)	)	-RRB-
2163	223	38	.	.	.
2163	224	1	11	11	CD
2163	224	2	.	.	.
2163	225	1	Yong	Yong	NNP
2163	225	2	Kiat	Kiat	NNP
2163	225	3	Ng	Ng	NNP
2163	225	4	,	,	,
2163	225	5	“	"	``
2163	225	6	Citation	Citation	NNP
2163	225	7	Parsing	Parsing	NNP
2163	225	8	Using	use	VBG
2163	225	9	Maximum	Maximum	NNP
2163	225	10	Entropy	Entropy	NNP
2163	225	11	and	and	CC
2163	225	12	Repairs	Repairs	NNPS
2163	225	13	”	"	''
2163	225	14	(	(	-LRB-
2163	225	15	master	master	NNP
2163	225	16	's	's	POS
2163	225	17	thesis	thesis	NN
2163	225	18	,	,	,
2163	225	19	National	National	NNP
2163	225	20	University	University	NNP
2163	225	21	of	of	IN
2163	225	22	Singapore	Singapore	NNP
2163	225	23	,	,	,
2163	225	24	2004	2004	CD
2163	225	25	)	)	-RRB-
2163	225	26	.	.	.
2163	226	1	12	12	CD
2163	226	2	.	.	.
2163	227	1	Fuchun	Fuchun	NNP
2163	227	2	Peng	Peng	NNP
2163	227	3	and	and	CC
2163	227	4	Andrew	Andrew	NNP
2163	227	5	McCallum	McCallum	NNP
2163	227	6	,	,	,
2163	227	7	“	"	``
2163	227	8	Information	Information	NNP
2163	227	9	Extraction	Extraction	NNP
2163	227	10	from	from	IN
2163	227	11	Research	Research	NNP
2163	227	12	Papers	Papers	NNPS
2163	227	13	Using	use	VBG
2163	227	14	Conditional	Conditional	NNP
2163	227	15	Random	Random	NNP
2163	227	16	Fields	Fields	NNPS
2163	227	17	,	,	,
2163	227	18	”	"	''
2163	227	19	Information	Information	NNP
2163	227	20	Processing	Processing	NNP
2163	227	21	&	&	CC
2163	227	22	Management	Management	NNP
2163	227	23	42	42	CD
2163	227	24	(	(	-LRB-
2163	227	25	2006	2006	CD
2163	227	26	)	)	-RRB-
2163	227	27	:	:	:
2163	227	28	963–79	963–79	CD
2163	227	29	.	.	.
2163	228	1	13	13	CD
2163	228	2	.	.	.
2163	229	1	C.	C.	NNP
2163	229	2	Lee	Lee	NNP
2163	229	3	Giles	Giles	NNP
2163	229	4	et	et	NNP
2163	229	5	al	al	NNP
2163	229	6	.	.	NNP
2163	229	7	,	,	,
2163	229	8	“	"	``
2163	229	9	CiteSeer	CiteSeer	NNP
2163	229	10	:	:	:
2163	229	11	An	an	DT
2163	229	12	Automatic	Automatic	NNP
2163	229	13	Citation	Citation	NNP
2163	229	14	Indexing	Indexing	NNP
2163	229	15	System	System	NNP
2163	229	16	,	,	,
2163	229	17	”	"	''
2163	229	18	Proceedings	proceeding	NNS
2163	229	19	of	of	IN
2163	229	20	the	the	DT
2163	229	21	Third	Third	NNP
2163	229	22	AMC	AMC	NNP
2163	229	23	Conference	Conference	NNP
2163	229	24	on	on	IN
2163	229	25	Digital	Digital	NNP
2163	229	26	Libraries	Libraries	NNP
2163	229	27	(	(	-LRB-
2163	229	28	Pittsburgh	Pittsburgh	NNP
2163	229	29	,	,	,
2163	229	30	PA	PA	NNP
2163	229	31	,	,	,
2163	229	32	1998	1998	CD
2163	229	33	)	)	-RRB-
2163	229	34	:	:	:
2163	229	35	89–98	89–98	LS
2163	229	36	.	.	.
2163	230	1	14	14	CD
2163	230	2	.	.	.
2163	231	1	Kristie	Kristie	NNP
2163	231	2	Seymore	Seymore	NNP
2163	231	3	et	et	NNP
2163	231	4	al	al	NNP
2163	231	5	.	.	NNP
2163	231	6	,	,	,
2163	231	7	“	"	``
2163	231	8	Learning	learn	VBG
2163	231	9	Hidden	Hidden	NNP
2163	231	10	Markov	Markov	NNP
2163	231	11	Model	Model	NNP
2163	231	12	Structure	Structure	NNP
2163	231	13	for	for	IN
2163	231	14	Information	Information	NNP
2163	231	15	Extraction	Extraction	NNP
2163	231	16	,	,	,
2163	231	17	”	"	''
2163	231	18	Proceedings	proceeding	NNS
2163	231	19	of	of	IN
2163	231	20	the	the	DT
2163	231	21	AAAI	AAAI	NNP
2163	231	22	Workshop	Workshop	NNP
2163	231	23	on	on	IN
2163	231	24	Machine	Machine	NNP
2163	231	25	Learning	Learning	NNP
2163	231	26	for	for	IN
2163	231	27	Information	Information	NNP
2163	231	28	Extraction	Extraction	NNP
2163	231	29	(	(	-LRB-
2163	231	30	1999	1999	CD
2163	231	31	)	)	-RRB-
2163	231	32	:	:	:
2163	231	33	37	37	CD
2163	231	34	–	–	:
2163	231	35	42	42	CD
2163	231	36	.	.	.
2163	232	1	15	15	CD
2163	232	2	.	.	.
2163	233	1	Isaac	Isaac	NNP
2163	233	2	G.	G.	NNP
2163	233	3	Councill	Councill	NNP
2163	233	4	et	et	FW
2163	233	5	al	al	NNP
2163	233	6	.	.	NNP
2163	233	7	,	,	,
2163	233	8	“	"	``
2163	233	9	ParsCit	parscit	NN
2163	233	10	:	:	:
2163	233	11	An	an	DT
2163	233	12	Open	open	JJ
2163	233	13	-	-	HYPH
2163	233	14	source	source	NN
2163	233	15	CRF	CRF	NNP
2163	233	16	Reference	Reference	NNP
2163	233	17	String	String	NNP
2163	233	18	Parsing	Parsing	NNP
2163	233	19	Package	package	NN
2163	233	20	,	,	,
2163	233	21	”	"	''
2163	233	22	Proceedings	proceeding	NNS
2163	233	23	of	of	IN
2163	233	24	the	the	DT
2163	233	25	Sixth	Sixth	NNP
2163	233	26	International	International	NNP
2163	233	27	Language	Language	NNP
2163	233	28	Resources	Resources	NNPS
2163	233	29	and	and	CC
2163	233	30	Evaluation	Evaluation	NNP
2163	233	31	(	(	-LRB-
2163	233	32	Marrakech	Marrakech	NNP
2163	233	33	,	,	,
2163	233	34	Morocco	Morocco	NNP
2163	233	35	,	,	,
2163	233	36	2008	2008	CD
2163	233	37	)	)	-RRB-
2163	233	38	.	.	.
2163	234	1	16	16	CD
2163	234	2	.	.	.
2163	235	1	Hui	Hui	NNP
2163	235	2	Han	Han	NNP
2163	235	3	et	et	NNP
2163	235	4	al	al	NNP
2163	235	5	.	.	NNP
2163	235	6	,	,	,
2163	235	7	“	"	``
2163	235	8	Rule	rule	NN
2163	235	9	-	-	HYPH
2163	235	10	based	base	VBN
2163	235	11	Word	Word	NNP
2163	235	12	Clustering	Clustering	NNP
2163	235	13	for	for	IN
2163	235	14	Document	document	NN
2163	235	15	Metadata	Metadata	NNP
2163	235	16	Extraction	Extraction	NNP
2163	235	17	,	,	,
2163	235	18	”	"	''
2163	235	19	Proceedings	proceeding	NNS
2163	235	20	of	of	IN
2163	235	21	the	the	DT
2163	235	22	Symposium	Symposium	NNP
2163	235	23	on	on	IN
2163	235	24	Applied	Applied	NNP
2163	235	25	Computing	Computing	NNP
2163	235	26	(	(	-LRB-
2163	235	27	Santa	Santa	NNP
2163	235	28	Fe	Fe	NNP
2163	235	29	,	,	,
2163	235	30	New	New	NNP
2163	235	31	Mexico	Mexico	NNP
2163	235	32	,	,	,
2163	235	33	2005	2005	CD
2163	235	34	)	)	-RRB-
2163	235	35	.	.	.
2163	236	1	17	17	CD
2163	236	2	.	.	.
2163	237	1	Eli	Eli	NNP
2163	237	2	Cortez	Cortez	NNPS
2163	237	3	et	et	NNP
2163	237	4	al	al	NNP
2163	237	5	.	.	NNP
2163	237	6	,	,	,
2163	237	7	“	"	``
2163	237	8	FLUX	FLUX	NNP
2163	237	9	-	-	HYPH
2163	237	10	CIM	CIM	NNP
2163	237	11	:	:	:
2163	237	12	Flexible	flexible	JJ
2163	237	13	Unsupervised	unsupervised	JJ
2163	237	14	Extraction	Extraction	NNP
2163	237	15	of	of	IN
2163	237	16	Citation	Citation	NNP
2163	237	17	Metadata	Metadata	NNP
2163	237	18	,	,	,
2163	237	19	”	"	''
2163	237	20	Proceedings	proceeding	NNS
2163	237	21	of	of	IN
2163	237	22	the	the	DT
2163	237	23	2007	2007	CD
2163	237	24	Conference	Conference	NNP
2163	237	25	on	on	IN
2163	237	26	Digital	Digital	NNP
2163	237	27	Libraries	Libraries	NNPS
2163	237	28	(	(	-LRB-
2163	237	29	New	New	NNP
2163	237	30	York	York	NNP
2163	237	31	,	,	,
2163	237	32	2007	2007	CD
2163	237	33	)	)	-RRB-
2163	237	34	:	:	:
2163	237	35	215–24	215–24	CD
2163	237	36	.	.	.
2163	238	1	18	18	CD
2163	238	2	.	.	.
2163	239	1	Machine	Machine	NNP
2163	239	2	Learning	Learning	NNP
2163	239	3	methods	method	NNS
2163	239	4	can	can	MD
2163	239	5	be	be	VB
2163	239	6	broadly	broadly	RB
2163	239	7	classified	classify	VBN
2163	239	8	into	into	IN
2163	239	9	two	two	CD
2163	239	10	categories	category	NNS
2163	239	11	:	:	:
2163	239	12	supervised	supervise	VBN
2163	239	13	and	and	CC
2163	239	14	unsupervised	unsupervised	JJ
2163	239	15	.	.	.
2163	240	1	Supervised	supervise	VBN
2163	240	2	methods	method	NNS
2163	240	3	require	require	VBP
2163	240	4	training	training	NN
2163	240	5	on	on	IN
2163	240	6	specific	specific	JJ
2163	240	7	datasets	dataset	NNS
2163	240	8	that	that	WDT
2163	240	9	exhibit	exhibit	VBP
2163	240	10	the	the	DT
2163	240	11	characteristics	characteristic	NNS
2163	240	12	of	of	IN
2163	240	13	the	the	DT
2163	240	14	target	target	NN
2163	240	15	domain	domain	NN
2163	240	16	.	.	.
2163	241	1	To	to	TO
2163	241	2	achieve	achieve	VB
2163	241	3	high	high	JJ
2163	241	4	accuracy	accuracy	NN
2163	241	5	levels	level	NNS
2163	241	6	,	,	,
2163	241	7	the	the	DT
2163	241	8	training	training	NN
2163	241	9	dataset	dataset	NN
2163	241	10	needs	need	VBZ
2163	241	11	to	to	TO
2163	241	12	be	be	VB
2163	241	13	reasonably	reasonably	RB
2163	241	14	large	large	JJ
2163	241	15	,	,	,
2163	241	16	and	and	CC
2163	241	17	more	more	RBR
2163	241	18	importantly	importantly	RB
2163	241	19	,	,	,
2163	241	20	it	-PRON-	PRP
2163	241	21	has	have	VBZ
2163	241	22	to	to	TO
2163	241	23	cover	cover	VB
2163	241	24	most	most	JJS
2163	241	25	of	of	IN
2163	241	26	the	the	DT
2163	241	27	possible	possible	JJ
2163	241	28	INFORMATION	information	JJ
2163	241	29	TECHNOLOGY	technology	NN
2163	241	30	AND	and	CC
2163	241	31	LIBRARIES	LIBRARIES	NNP
2163	241	32	|	|	NNP
2163	241	33	JUNE	JUNE	NNP
2163	241	34	2012	2012	CD
2163	241	35	19	19	CD
2163	241	36	exceptions	exception	NNS
2163	241	37	from	from	IN
2163	241	38	the	the	DT
2163	241	39	intrinsic	intrinsic	JJ
2163	241	40	data	datum	NNS
2163	241	41	patterns	pattern	NNS
2163	241	42	.	.	.
2163	242	1	Unlike	unlike	IN
2163	242	2	supervised	supervise	VBN
2163	242	3	methods	method	NNS
2163	242	4	,	,	,
2163	242	5	unsupervised	unsupervised	JJ
2163	242	6	methods	method	NNS
2163	242	7	do	do	VBP
2163	242	8	not	not	RB
2163	242	9	require	require	VB
2163	242	10	training	training	NN
2163	242	11	,	,	,
2163	242	12	and	and	CC
2163	242	13	in	in	IN
2163	242	14	principle	principle	NN
2163	242	15	,	,	,
2163	242	16	use	use	VB
2163	242	17	generic	generic	JJ
2163	242	18	rules	rule	NNS
2163	242	19	to	to	TO
2163	242	20	encode	encode	VB
2163	242	21	both	both	DT
2163	242	22	the	the	DT
2163	242	23	expected	expected	JJ
2163	242	24	patterns	pattern	NNS
2163	242	25	and	and	CC
2163	242	26	the	the	DT
2163	242	27	possible	possible	JJ
2163	242	28	exceptions	exception	NNS
2163	242	29	of	of	IN
2163	242	30	the	the	DT
2163	242	31	target	target	NN
2163	242	32	data	datum	NNS
2163	242	33	.	.	.
2163	243	1	19	19	CD
2163	243	2	.	.	.
2163	244	1	Peng	Peng	NNP
2163	244	2	and	and	CC
2163	244	3	McCallum	McCallum	NNP
2163	244	4	,	,	,
2163	244	5	“	"	``
2163	244	6	Information	Information	NNP
2163	244	7	Extraction	Extraction	NNP
2163	244	8	from	from	IN
2163	244	9	Research	Research	NNP
2163	244	10	Papers	Papers	NNPS
2163	244	11	Using	use	VBG
2163	244	12	Conditional	Conditional	NNP
2163	244	13	Random	Random	NNP
2163	244	14	Fields	Fields	NNPS
2163	244	15	.	.	.
2163	244	16	”	"	''
2163	244	17	20	20	CD
2163	244	18	.	.	.
2163	245	1	Hoifung	Hoifung	NNP
2163	245	2	Poon	Poon	NNP
2163	245	3	and	and	CC
2163	245	4	Pedro	Pedro	NNP
2163	245	5	Domingos	Domingos	NNP
2163	245	6	,	,	,
2163	245	7	“	"	``
2163	245	8	Joint	joint	JJ
2163	245	9	inference	inference	NN
2163	245	10	in	in	IN
2163	245	11	information	information	NN
2163	245	12	extraction	extraction	NN
2163	245	13	,	,	,
2163	245	14	”	"	''
2163	245	15	Proceedings	proceeding	NNS
2163	245	16	of	of	IN
2163	245	17	the	the	DT
2163	245	18	22nd	22nd	JJ
2163	245	19	National	National	NNP
2163	245	20	Conference	Conference	NNP
2163	245	21	on	on	IN
2163	245	22	Artificial	Artificial	NNP
2163	245	23	Intelligence	Intelligence	NNP
2163	245	24	(	(	-LRB-
2163	245	25	Vancouver	Vancouver	NNP
2163	245	26	,	,	,
2163	245	27	British	British	NNP
2163	245	28	Columbia	Columbia	NNP
2163	245	29	,	,	,
2163	245	30	Canada	Canada	NNP
2163	245	31	,	,	,
2163	245	32	2007	2007	CD
2163	245	33	)	)	-RRB-
2163	245	34	:	:	:
2163	245	35	913–18	913–18	CD
2163	245	36	.	.	.
2163	246	1	21	21	CD
2163	246	2	.	.	.
2163	247	1	Ariel	Ariel	NNP
2163	247	2	Schwartz	Schwartz	NNP
2163	247	3	et	et	FW
2163	247	4	al	al	NNP
2163	247	5	.	.	NNP
2163	247	6	,	,	,
2163	247	7	“	"	``
2163	247	8	Multiple	Multiple	NNP
2163	247	9	Alignment	Alignment	NNP
2163	247	10	of	of	IN
2163	247	11	Citation	Citation	NNP
2163	247	12	Sentences	Sentences	NNPS
2163	247	13	with	with	IN
2163	247	14	Conditional	Conditional	NNP
2163	247	15	Random	Random	NNP
2163	247	16	Fields	Fields	NNPS
2163	247	17	and	and	CC
2163	247	18	Posterior	Posterior	NNP
2163	247	19	Decoding	decode	VBG
2163	247	20	,	,	,
2163	247	21	”	"	''
2163	247	22	Proceedings	proceeding	NNS
2163	247	23	of	of	IN
2163	247	24	the	the	DT
2163	247	25	2007	2007	CD
2163	247	26	Joint	Joint	NNP
2163	247	27	Conference	Conference	NNP
2163	247	28	on	on	IN
2163	247	29	Empirical	Empirical	NNP
2163	247	30	Methods	Methods	NNPS
2163	247	31	in	in	IN
2163	247	32	Natural	Natural	NNP
2163	247	33	Language	Language	NNP
2163	247	34	Processing	Processing	NNP
2163	247	35	and	and	CC
2163	247	36	Computational	Computational	NNP
2163	247	37	Natural	Natural	NNP
2163	247	38	Language	Language	NNP
2163	247	39	Learning	Learning	NNP
2163	247	40	(	(	-LRB-
2163	247	41	Prague	Prague	NNP
2163	247	42	,	,	,
2163	247	43	Czech	Czech	NNP
2163	247	44	Republic	Republic	NNP
2163	247	45	,	,	,
2163	247	46	2007	2007	CD
2163	247	47	)	)	-RRB-
2163	247	48	:	:	:
2163	247	49	847–57	847–57	CD
2163	247	50	.	.	.
2163	248	1	22	22	CD
2163	248	2	.	.	.
2163	249	1	Simone	Simone	NNP
2163	249	2	Teufel	Teufel	NNP
2163	249	3	et	et	NNP
2163	249	4	al	al	NNP
2163	249	5	.	.	NNP
2163	249	6	,	,	,
2163	249	7	“	"	``
2163	249	8	Automatic	automatic	JJ
2163	249	9	Classification	Classification	NNP
2163	249	10	of	of	IN
2163	249	11	Citation	Citation	NNP
2163	249	12	Function	Function	NNP
2163	249	13	,	,	,
2163	249	14	”	"	''
2163	249	15	Proceedings	proceeding	NNS
2163	249	16	of	of	IN
2163	249	17	the	the	DT
2163	249	18	2006	2006	CD
2163	249	19	Conference	Conference	NNP
2163	249	20	on	on	IN
2163	249	21	Empirical	Empirical	NNP
2163	249	22	Methods	Methods	NNPS
2163	249	23	in	in	IN
2163	249	24	Natural	Natural	NNP
2163	249	25	Language	Language	NNP
2163	249	26	Processing	Processing	NNP
2163	249	27	(	(	-LRB-
2163	249	28	Sydney	Sydney	NNP
2163	249	29	,	,	,
2163	249	30	Australia	Australia	NNP
2163	249	31	,	,	,
2163	249	32	2006	2006	CD
2163	249	33	)	)	-RRB-
2163	249	34	:	:	:
2163	249	35	103–10	103–10	CD
2163	249	36	.	.	.
2163	250	1	23	23	CD
2163	250	2	.	.	.
2163	251	1	Jien	Jien	NNP
2163	251	2	-	-	HYPH
2163	251	3	Chen	Chen	NNP
2163	251	4	Wu	Wu	NNP
2163	251	5	et	et	NNP
2163	251	6	al	al	NNP
2163	251	7	.	.	NNP
2163	251	8	,	,	,
2163	251	9	“	"	``
2163	251	10	Computational	Computational	NNP
2163	251	11	Analysis	Analysis	NNP
2163	251	12	of	of	IN
2163	251	13	Move	Move	NNP
2163	251	14	Structures	structure	NNS
2163	251	15	in	in	IN
2163	251	16	Academic	Academic	NNP
2163	251	17	Abstracts	Abstracts	NNPS
2163	251	18	,	,	,
2163	251	19	”	"	''
2163	251	20	COLING	COLING	NNP
2163	251	21	/	/	SYM
2163	251	22	ACL	ACL	NNP
2163	251	23	Interactive	Interactive	NNP
2163	251	24	Presentation	Presentation	NNP
2163	251	25	Sessions	Sessions	NNP
2163	251	26	(	(	-LRB-
2163	251	27	Sydney	Sydney	NNP
2163	251	28	,	,	,
2163	251	29	Australia	Australia	NNP
2163	251	30	,	,	,
2163	251	31	2006	2006	CD
2163	251	32	)	)	-RRB-
2163	251	33	:	:	:
2163	251	34	41–44	41–44	CD
2163	251	35	.	.	.
2163	252	1	24	24	CD
2163	252	2	.	.	.
2163	253	1	Eugenio	Eugenio	NNP
2163	253	2	Cesario	Cesario	NNP
2163	253	3	et	et	NNP
2163	253	4	al	al	NNP
2163	253	5	.	.	NNP
2163	253	6	,	,	,
2163	253	7	“	"	``
2163	253	8	Boosting	boost	VBG
2163	253	9	text	text	NN
2163	253	10	segmentation	segmentation	NN
2163	253	11	via	via	IN
2163	253	12	progressive	progressive	JJ
2163	253	13	classification	classification	NN
2163	253	14	,	,	,
2163	253	15	”	"	''
2163	253	16	Knowledge	knowledge	NN
2163	253	17	and	and	CC
2163	253	18	Information	Information	NNP
2163	253	19	Systems	Systems	NNP
2163	253	20	15	15	CD
2163	253	21	(	(	-LRB-
2163	253	22	2008	2008	CD
2163	253	23	)	)	-RRB-
2163	253	24	:	:	:
2163	253	25	285–320	285–320	CD
2163	253	26	.	.	.
2163	254	1	25	25	CD
2163	254	2	.	.	.
2163	255	1	Dublin	Dublin	NNP
2163	255	2	Core	Core	NNP
2163	255	3	website	website	NN
2163	255	4	,	,	,
2163	255	5	http://dublincore.org	http://dublincore.org	NNP
2163	255	6	(	(	-LRB-
2163	255	7	accessed	access	VBN
2163	255	8	May	May	NNP
2163	255	9	4	4	CD
2163	255	10	,	,	,
2163	255	11	2011	2011	CD
2163	255	12	)	)	-RRB-
2163	255	13	.	.	.
2163	256	1	26	26	CD
2163	256	2	.	.	.
2163	257	1	York	York	NNP
2163	257	2	Sure	sure	UH
2163	257	3	et	et	FW
2163	257	4	al	al	NNP
2163	257	5	.	.	NNP
2163	257	6	,	,	,
2163	257	7	“	"	``
2163	257	8	The	the	DT
2163	257	9	SWRC	SWRC	NNP
2163	257	10	ontology	ontology	NN
2163	257	11	–	–	:
2163	257	12	Semantic	Semantic	NNP
2163	257	13	Web	Web	NNP
2163	257	14	for	for	IN
2163	257	15	research	research	NN
2163	257	16	communities	community	NNS
2163	257	17	,	,	,
2163	257	18	”	"	''
2163	257	19	Proceedings	proceeding	NNS
2163	257	20	of	of	IN
2163	257	21	the	the	DT
2163	257	22	12th	12th	JJ
2163	257	23	Portuguese	Portuguese	NNP
2163	257	24	Conference	Conference	NNP
2163	257	25	on	on	IN
2163	257	26	Artificial	Artificial	NNP
2163	257	27	Intelligence	Intelligence	NNP
2163	257	28	(	(	-LRB-
2163	257	29	Covilha	Covilha	NNP
2163	257	30	,	,	,
2163	257	31	Portugal	Portugal	NNP
2163	257	32	,	,	,
2163	257	33	2005	2005	CD
2163	257	34	)	)	-RRB-
2163	257	35	.	.	.
2163	258	1	27	27	CD
2163	258	2	.	.	.
2163	259	1	Yanjun	Yanjun	NNP
2163	259	2	Qi	Qi	NNP
2163	259	3	et	et	NNP
2163	259	4	al	al	NNP
2163	259	5	.	.	NNP
2163	259	6	,	,	,
2163	259	7	“	"	``
2163	259	8	Semi	Semi	NNP
2163	259	9	-	-	HYPH
2163	259	10	Supervised	supervised	JJ
2163	259	11	Sequence	Sequence	NNP
2163	259	12	Labeling	labeling	NN
2163	259	13	with	with	IN
2163	259	14	Self	self	NN
2163	259	15	-	-	HYPH
2163	259	16	Learned	learn	VBN
2163	259	17	Features	feature	NNS
2163	259	18	,	,	,
2163	259	19	”	"	''
2163	259	20	Proceedings	proceeding	NNS
2163	259	21	of	of	IN
2163	259	22	IEEE	IEEE	NNP
2163	259	23	International	International	NNP
2163	259	24	Conference	Conference	NNP
2163	259	25	on	on	IN
2163	259	26	Data	Data	NNP
2163	259	27	Mining	Mining	NNP
2163	259	28	(	(	-LRB-
2163	259	29	Miami	Miami	NNP
2163	259	30	,	,	,
2163	259	31	FL	FL	NNP
2163	259	32	,	,	,
2163	259	33	USA	USA	NNP
2163	259	34	,	,	,
2163	259	35	2009	2009	CD
2163	259	36	)	)	-RRB-
2163	259	37	.	.	.
2163	260	1	28	28	CD
2163	260	2	.	.	.
2163	261	1	David	David	NNP
2163	261	2	Sanchez	Sanchez	NNP
2163	261	3	et	et	NNP
2163	261	4	al	al	NNP
2163	261	5	.	.	NNP
2163	261	6	,	,	,
2163	261	7	“	"	``
2163	261	8	Content	content	NN
2163	261	9	Annotation	annotation	NN
2163	261	10	for	for	IN
2163	261	11	the	the	DT
2163	261	12	Semantic	semantic	JJ
2163	261	13	Web	web	NN
2163	261	14	:	:	:
2163	261	15	An	an	DT
2163	261	16	Automatic	Automatic	NNP
2163	261	17	Web	web	NN
2163	261	18	-	-	HYPH
2163	261	19	Based	base	VBN
2163	261	20	Approach	Approach	NNP
2163	261	21	,	,	,
2163	261	22	”	"	''
2163	261	23	Knowledge	knowledge	NN
2163	261	24	and	and	CC
2163	261	25	Information	Information	NNP
2163	261	26	Systems	Systems	NNP
2163	261	27	27	27	CD
2163	261	28	(	(	-LRB-
2163	261	29	2011	2011	CD
2163	261	30	)	)	-RRB-
2163	261	31	:	:	:
2163	261	32	393	393	CD
2163	261	33	-	-	SYM
2163	261	34	418	418	CD
2163	261	35	.	.	.
2163	262	1	29	29	CD
2163	262	2	.	.	.
2163	263	1	Tshering	tshere	VBG
2163	263	2	Cigay	Cigay	NNP
2163	263	3	Dorji	Dorji	NNP
2163	263	4	et	et	NNP
2163	263	5	al	al	NNP
2163	263	6	.	.	NNP
2163	263	7	,	,	,
2163	263	8	“	"	``
2163	263	9	Extraction	extraction	NN
2163	263	10	,	,	,
2163	263	11	selection	selection	NN
2163	263	12	and	and	CC
2163	263	13	ranking	ranking	NN
2163	263	14	of	of	IN
2163	263	15	Field	Field	NNP
2163	263	16	Association	Association	NNP
2163	263	17	(	(	-LRB-
2163	263	18	FA	FA	NNP
2163	263	19	)	)	-RRB-
2163	263	20	Terms	term	NNS
2163	263	21	from	from	IN
2163	263	22	domain	domain	NN
2163	263	23	-	-	HYPH
2163	263	24	specific	specific	JJ
2163	263	25	corpora	corpora	NN
2163	263	26	for	for	IN
2163	263	27	building	build	VBG
2163	263	28	a	a	DT
2163	263	29	comprehensive	comprehensive	JJ
2163	263	30	FA	FA	NNP
2163	263	31	terms	term	NNS
2163	263	32	dictionary	dictionary	JJ
2163	263	33	,	,	,
2163	263	34	”	"	''
2163	263	35	Knowledge	Knowledge	NNP
2163	263	36	and	and	CC
2163	263	37	Information	Information	NNP
2163	263	38	Systems	Systems	NNP
2163	263	39	27	27	CD
2163	263	40	(	(	-LRB-
2163	263	41	2011	2011	CD
2163	263	42	)	)	-RRB-
2163	263	43	:	:	:
2163	263	44	141–61	141–61	CD
2163	263	45	.	.	.
2163	264	1	30	30	CD
2163	264	2	.	.	.
2163	265	1	Please	please	UH
2163	265	2	note	note	VB
2163	265	3	that	that	IN
2163	265	4	the	the	DT
2163	265	5	chunker	chunker	NN
2163	265	6	is	be	VBZ
2163	265	7	document	document	NN
2163	265	8	-	-	HYPH
2163	265	9	format	format	NN
2163	265	10	agnostic	agnostic	JJ
2163	265	11	and	and	CC
2163	265	12	takes	take	VBZ
2163	265	13	as	as	RB
2163	265	14	input	input	NN
2163	265	15	only	only	JJ
2163	265	16	raw	raw	JJ
2163	265	17	text	text	NN
2163	265	18	.	.	.
2163	266	1	The	the	DT
2163	266	2	actual	actual	JJ
2163	266	3	extraction	extraction	NN
2163	266	4	of	of	IN
2163	266	5	this	this	DT
2163	266	6	raw	raw	JJ
2163	266	7	text	text	NN
2163	266	8	from	from	IN
2163	266	9	the	the	DT
2163	266	10	original	original	JJ
2163	266	11	document	document	NN
2163	266	12	(	(	-LRB-
2163	266	13	PDF	PDF	NNP
2163	266	14	,	,	,
2163	266	15	DOC	DOC	NNP
2163	266	16	or	or	CC
2163	266	17	some	some	DT
2163	266	18	other	other	JJ
2163	266	19	format	format	NN
2163	266	20	)	)	-RRB-
2163	266	21	is	be	VBZ
2163	266	22	the	the	DT
2163	266	23	user	user	NN
2163	266	24	’s	’s	POS
2163	266	25	responsibility	responsibility	NN
2163	266	26	.	.	.
2163	267	1	31	31	CD
2163	267	2	.	.	.
2163	268	1	As	as	IN
2163	268	2	a	a	DT
2163	268	3	note	note	NN
2163	268	4	,	,	,
2163	268	5	we	-PRON-	PRP
2163	268	6	chose	choose	VBD
2163	268	7	this	this	DT
2163	268	8	length	length	NN
2163	268	9	of	of	IN
2163	268	10	fifteen	fifteen	CD
2163	268	11	characters	character	NNS
2163	268	12	empirically	empirically	RB
2163	268	13	,	,	,
2163	268	14	and	and	CC
2163	268	15	based	base	VBN
2163	268	16	on	on	IN
2163	268	17	the	the	DT
2163	268	18	assumption	assumption	NN
2163	268	19	that	that	IN
2163	268	20	in	in	IN
2163	268	21	any	any	DT
2163	268	22	format	format	NN
2163	268	23	the	the	DT
2163	268	24	publication	publication	NN
2163	268	25	content	content	NN
2163	268	26	lines	line	NNS
2163	268	27	usually	usually	RB
2163	268	28	have	have	VBP
2163	268	29	more	more	JJR
2163	268	30	than	than	IN
2163	268	31	fifteen	fifteen	CD
2163	268	32	characters	character	NNS
2163	268	33	.	.	.
2163	269	1	REFERENCE	reference	NN
2163	269	2	INFORMATION	INFORMATION	NNS
2163	269	3	EXTRACTION	extraction	NN
2163	269	4	AND	and	CC
2163	269	5	PROCESSING	processing	NN
2163	269	6	|GROZA	|groza	NN
2163	269	7	,	,	,
2163	269	8	GRIMNES	GRIMNES	NNP
2163	269	9	,	,	,
2163	269	10	AND	and	CC
2163	269	11	HANDSCHUH	handschuh	NN
2163	269	12	20	20	CD
2163	269	13	32	32	CD
2163	269	14	.	.	.
2163	270	1	Lafferty	Lafferty	NNP
2163	270	2	et	et	NNP
2163	270	3	al	al	NNP
2163	270	4	.	.	NNP
2163	270	5	,	,	,
2163	270	6	“	"	``
2163	270	7	Conditional	Conditional	NNP
2163	270	8	Random	Random	NNP
2163	270	9	Fields	Fields	NNPS
2163	270	10	:	:	:
2163	270	11	Probabilistic	Probabilistic	NNP
2163	270	12	Models	Models	NNPS
2163	270	13	for	for	IN
2163	270	14	Segmenting	segmenting	NN
2163	270	15	and	and	CC
2163	270	16	Labeling	Labeling	NNP
2163	270	17	Sequence	Sequence	NNP
2163	270	18	Data	Data	NNP
2163	270	19	.	.	.
2163	270	20	”	"	''
2163	270	21	33	33	CD
2163	270	22	.	.	.
2163	271	1	Councill	Councill	NNP
2163	271	2	et	et	FW
2163	271	3	al	al	NNP
2163	271	4	.	.	NNP
2163	271	5	,	,	,
2163	271	6	“	"	``
2163	271	7	ParsCit	parscit	NN
2163	271	8	:	:	:
2163	271	9	An	an	DT
2163	271	10	Open	open	JJ
2163	271	11	-	-	HYPH
2163	271	12	source	source	NN
2163	271	13	CRF	CRF	NNP
2163	271	14	Reference	Reference	NNP
2163	271	15	String	String	NNP
2163	271	16	Parsing	Parsing	NNP
2163	271	17	Package	Package	NNP
2163	271	18	.	.	.
2163	271	19	”	"	''
2163	271	20	34	34	CD
2163	271	21	.	.	.
2163	272	1	The	the	DT
2163	272	2	manual	manual	JJ
2163	272	3	tagging	tagging	NN
2163	272	4	was	be	VBD
2163	272	5	performed	perform	VBN
2163	272	6	by	by	IN
2163	272	7	a	a	DT
2163	272	8	single	single	JJ
2163	272	9	person	person	NN
2163	272	10	and	and	CC
2163	272	11	since	since	IN
2163	272	12	the	the	DT
2163	272	13	reference	reference	NN
2163	272	14	chunks	chunk	NNS
2163	272	15	have	have	VBP
2163	272	16	no	no	DT
2163	272	17	ambiguity	ambiguity	NN
2163	272	18	attached	attach	VBN
2163	272	19	,	,	,
2163	272	20	we	-PRON-	PRP
2163	272	21	did	do	VBD
2163	272	22	not	not	RB
2163	272	23	see	see	VB
2163	272	24	the	the	DT
2163	272	25	need	need	NN
2163	272	26	for	for	IN
2163	272	27	running	run	VBG
2163	272	28	any	any	DT
2163	272	29	data	data	NN
2163	272	30	reliability	reliability	NN
2163	272	31	tests	test	NNS
2163	272	32	.	.	.
2163	273	1	35	35	CD
2163	273	2	.	.	.
2163	274	1	Ron	Ron	NNP
2163	274	2	Kohavi	Kohavi	NNP
2163	274	3	,	,	,
2163	274	4	“	"	``
2163	274	5	A	a	DT
2163	274	6	Study	Study	NNP
2163	274	7	of	of	IN
2163	274	8	Cross	Cross	NNP
2163	274	9	-	-	NNP
2163	274	10	Validation	Validation	NNP
2163	274	11	and	and	CC
2163	274	12	Bootstrap	Bootstrap	NNP
2163	274	13	for	for	IN
2163	274	14	Accuracy	Accuracy	NNP
2163	274	15	Estimation	Estimation	NNP
2163	274	16	and	and	CC
2163	274	17	Model	Model	NNP
2163	274	18	Selection	Selection	NNP
2163	274	19	,	,	,
2163	274	20	”	"	''
2163	274	21	Proceedings	proceeding	NNS
2163	274	22	of	of	IN
2163	274	23	the	the	DT
2163	274	24	14th	14th	JJ
2163	274	25	International	International	NNP
2163	274	26	Joint	Joint	NNP
2163	274	27	Conference	Conference	NNP
2163	274	28	on	on	IN
2163	274	29	Artificial	Artificial	NNP
2163	274	30	Intelligence	Intelligence	NNP
2163	274	31	(	(	-LRB-
2163	274	32	Montreal	Montreal	NNP
2163	274	33	,	,	,
2163	274	34	Quebec	Quebec	NNP
2163	274	35	,	,	,
2163	274	36	1995	1995	CD
2163	274	37	)	)	-RRB-
2163	274	38	:	:	:
2163	274	39	1137–43	1137–43	LS
2163	274	40	.	.	.
2163	275	1	36	36	CD
2163	275	2	.	.	.
2163	276	1	Peng	Peng	NNP
2163	276	2	and	and	CC
2163	276	3	McCallum	McCallum	NNP
2163	276	4	,	,	,
2163	276	5	“	"	``
2163	276	6	Information	Information	NNP
2163	276	7	Extraction	Extraction	NNP
2163	276	8	from	from	IN
2163	276	9	Research	Research	NNP
2163	276	10	Papers	Papers	NNPS
2163	276	11	Using	use	VBG
2163	276	12	Conditional	Conditional	NNP
2163	276	13	Random	Random	NNP
2163	276	14	Fields	Fields	NNPS
2163	276	15	.	.	.
2163	276	16	”	"	''
2163	276	17	37	37	CD
2163	276	18	.	.	.
2163	277	1	Councill	Councill	NNP
2163	277	2	et	et	FW
2163	277	3	al	al	NNP
2163	277	4	.	.	NNP
2163	277	5	,	,	,
2163	277	6	“	"	``
2163	277	7	ParsCit	parscit	NN
2163	277	8	:	:	:
2163	277	9	An	an	DT
2163	277	10	Open	open	JJ
2163	277	11	-	-	HYPH
2163	277	12	source	source	NN
2163	277	13	CRF	CRF	NNP
2163	277	14	Reference	Reference	NNP
2163	277	15	String	String	NNP
2163	277	16	Parsing	Parsing	NNP
2163	277	17	Package	Package	NNP
2163	277	18	.	.	.
2163	277	19	”	"	''
2163	277	20	38	38	CD
2163	277	21	.	.	.
2163	278	1	Mallet	mallet	NN
2163	278	2	:	:	:
2163	278	3	MAchine	MAchine	NNP
2163	278	4	Learning	Learning	NNP
2163	278	5	for	for	IN
2163	278	6	LanguagE	LanguagE	NNP
2163	278	7	Toolkit	Toolkit	NNP
2163	278	8	,	,	,
2163	278	9	http://mallet.cs.umass.edu	http://mallet.cs.umass.edu	NNS
2163	278	10	(	(	-LRB-
2163	278	11	accessed	access	VBN
2163	278	12	May	May	NNP
2163	278	13	4	4	CD
2163	278	14	,	,	,
2163	278	15	2011	2011	CD
2163	278	16	)	)	-RRB-
2163	278	17	.	.	.
2163	279	1	39	39	CD
2163	279	2	.	.	.
2163	280	1	William	William	NNP
2163	280	2	M.	M.	NNP
2163	280	3	Shaw	Shaw	NNP
2163	280	4	et	et	FW
2163	280	5	al	al	NNP
2163	280	6	.	.	NNP
2163	280	7	,	,	,
2163	280	8	“	"	``
2163	280	9	Performance	performance	NN
2163	280	10	standards	standard	NNS
2163	280	11	and	and	CC
2163	280	12	evaluations	evaluation	NNS
2163	280	13	in	in	IN
2163	280	14	IR	IR	NNP
2163	280	15	test	test	NN
2163	280	16	collections	collection	NNS
2163	280	17	:	:	:
2163	280	18	Cluster-	Cluster-	NNP
2163	280	19	based	base	VBN
2163	280	20	retrieval	retrieval	NN
2163	280	21	models	model	NNS
2163	280	22	,	,	,
2163	280	23	”	"	''
2163	280	24	Information	Information	NNP
2163	280	25	Processing	Processing	NNP
2163	280	26	&	&	CC
2163	280	27	Management	Management	NNP
2163	280	28	33	33	CD
2163	280	29	(	(	-LRB-
2163	280	30	1997	1997	CD
2163	280	31	)	)	-RRB-
2163	280	32	:	:	:
2163	280	33	1–14	1–14	CD
2163	280	34	.	.	.
2163	281	1	40	40	CD
2163	281	2	.	.	.
2163	282	1	Peng	Peng	NNP
2163	282	2	and	and	CC
2163	282	3	McCallum	McCallum	NNP
2163	282	4	,	,	,
2163	282	5	“	"	``
2163	282	6	Information	Information	NNP
2163	282	7	Extraction	Extraction	NNP
2163	282	8	from	from	IN
2163	282	9	Research	Research	NNP
2163	282	10	Papers	Papers	NNPS
2163	282	11	Using	use	VBG
2163	282	12	Conditional	Conditional	NNP
2163	282	13	Random	Random	NNP
2163	282	14	Fields	Fields	NNPS
2163	282	15	.	.	.
2163	282	16	”	"	''
2163	282	17	41	41	CD
2163	282	18	.	.	.
2163	283	1	Councill	Councill	NNP
2163	283	2	et	et	FW
2163	283	3	al	al	NNP
2163	283	4	.	.	NNP
2163	283	5	,	,	,
2163	283	6	“	"	``
2163	283	7	ParsCit	parscit	NN
2163	283	8	:	:	:
2163	283	9	An	an	DT
2163	283	10	Open	open	JJ
2163	283	11	-	-	HYPH
2163	283	12	source	source	NN
2163	283	13	CRF	CRF	NNP
2163	283	14	Reference	Reference	NNP
2163	283	15	String	String	NNP
2163	283	16	Parsing	Parsing	NNP
2163	283	17	Package	Package	NNP
2163	283	18	.	.	.
2163	283	19	”	"	''
2163	283	20	42	42	CD
2163	283	21	.	.	.
2163	284	1	Seymore	Seymore	NNP
2163	284	2	et	et	NNP
2163	284	3	al	al	NNP
2163	284	4	.	.	NNP
2163	284	5	,	,	,
2163	284	6	“	"	``
2163	284	7	Learning	learn	VBG
2163	284	8	Hidden	Hidden	NNP
2163	284	9	Markov	Markov	NNP
2163	284	10	Model	Model	NNP
2163	284	11	Structure	Structure	NNP
2163	284	12	for	for	IN
2163	284	13	Information	Information	NNP
2163	284	14	Extraction	Extraction	NNP
2163	284	15	.	.	.
2163	284	16	”	"	''
2163	284	17	43	43	CD
2163	284	18	.	.	.
2163	285	1	Han	Han	NNP
2163	285	2	et	et	NNP
2163	285	3	al	al	NNP
2163	285	4	.	.	NNP
2163	285	5	,	,	,
2163	285	6	“	"	``
2163	285	7	Rule	rule	NN
2163	285	8	-	-	HYPH
2163	285	9	based	base	VBN
2163	285	10	Word	Word	NNP
2163	285	11	Clustering	Clustering	NNP
2163	285	12	for	for	IN
2163	285	13	Document	document	NN
2163	285	14	Metadata	Metadata	NNP
2163	285	15	Extraction	Extraction	NNP
2163	285	16	.	.	.
2163	285	17	”	"	''
2163	285	18	44	44	CD
2163	285	19	.	.	.
2163	286	1	Cortez	Cortez	NNP
2163	286	2	et	et	NNP
2163	286	3	al	al	NNP
2163	286	4	.	.	NNP
2163	286	5	,	,	,
2163	286	6	“	"	``
2163	286	7	FLUX	FLUX	NNP
2163	286	8	-	-	HYPH
2163	286	9	CIM	CIM	NNP
2163	286	10	:	:	:
2163	286	11	Flexible	flexible	JJ
2163	286	12	Unsupervised	unsupervised	JJ
2163	286	13	Extraction	Extraction	NNP
2163	286	14	of	of	IN
2163	286	15	Citation	Citation	NNP
2163	286	16	Metadata	Metadata	NNP
2163	286	17	.	.	.
2163	286	18	”	"	''
2163	286	19	45	45	CD
2163	286	20	.	.	.
2163	287	1	The	the	DT
2163	287	2	CS	CS	NNP
2163	287	3	-	-	HYPH
2163	287	4	SW	SW	NNP
2163	287	5	dataset	dataset	NN
2163	287	6	is	be	VBZ
2163	287	7	available	available	JJ
2163	287	8	at	at	IN
2163	287	9	http://resources.smile.deri.ie/corpora/cs-sw	http://resources.smile.deri.ie/corpora/cs-sw	ADD
2163	287	10	(	(	-LRB-
2163	287	11	accessed	access	VBN
2163	287	12	May	May	NNP
2163	287	13	4	4	CD
2163	287	14	,	,	,
2163	287	15	2011	2011	CD
2163	287	16	)	)	-RRB-
2163	287	17	.	.	.
2163	288	1	http://resources.smile.deri.ie/corpora/cs-sw	http://resources.smile.deri.ie/corpora/cs-sw	ADD