ON THE EFFECT OF WORD POSITIONS IN GRAPH-BASED KEYWORD EXTRACTION

In this study, we focus on the effect of word positions in unsupervised, graph-based keyword extraction. To this aim, we discuss the performance of four node-weighting procedures, namely Word Position (WP), Word Position Bidirectional (WPB), Sentence Position (SP), and Sentence Position Bidirectiona...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Osman KABASAKAL, Alev MUTLU
Formato:	article
Lenguaje:	EN
Publicado:	National Defense University Barbaros Naval Sciences and Engineering Institute Journal of Naval Science and Engineering 2021
Materias:	keyword extraction sentence position word position Naval Science V
Acceso en línea:	https://doaj.org/article/327c54751393413bbd6c9c650b41b4e6
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:327c54751393413bbd6c9c650b41b4e6
record_format	dspace
spelling	oai:doaj.org-article:327c54751393413bbd6c9c650b41b4e62021-11-16T15:40:56ZON THE EFFECT OF WORD POSITIONS IN GRAPH-BASED KEYWORD EXTRACTION1304-2025https://doaj.org/article/327c54751393413bbd6c9c650b41b4e62021-11-01T00:00:00Zhttps://dergipark.org.tr/tr/download/article-file/1420329https://doaj.org/toc/1304-2025In this study, we focus on the effect of word positions in unsupervised, graph-based keyword extraction. To this aim, we discuss the performance of four node-weighting procedures, namely Word Position (WP), Word Position Bidirectional (WPB), Sentence Position (SP), and Sentence Position Bidirectional (SPB). WP assigns higher weights to words that appear at the beginning of a text. WPB assigns higher weights to words that appear either at the beginning or end of a text. SP assigns higher weights to words that appear in the very first sentences of a text. SPB assigns higher weights to words that appear in sentences that are either close to the beginning or end of a text. Experiments conducted on six benchmark datasets show that WP and SP do not statistically differ. However, for datasets whose keywords appear early in the text WP performs better than SP with no statistical difference, while for datasets where keywords are evenly distributed in text SP statistically performs better than WP.Osman KABASAKALAlev MUTLUNational Defense University Barbaros Naval Sciences and Engineering Institute Journal of Naval Science and Engineeringarticlekeyword extractionsentence positionword positionNaval ScienceVENJournal of Naval Science and Engineering, Vol 17, Iss 2, Pp 217-239 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	keyword extraction sentence position word position Naval Science V
spellingShingle	keyword extraction sentence position word position Naval Science V Osman KABASAKAL Alev MUTLU ON THE EFFECT OF WORD POSITIONS IN GRAPH-BASED KEYWORD EXTRACTION
description	In this study, we focus on the effect of word positions in unsupervised, graph-based keyword extraction. To this aim, we discuss the performance of four node-weighting procedures, namely Word Position (WP), Word Position Bidirectional (WPB), Sentence Position (SP), and Sentence Position Bidirectional (SPB). WP assigns higher weights to words that appear at the beginning of a text. WPB assigns higher weights to words that appear either at the beginning or end of a text. SP assigns higher weights to words that appear in the very first sentences of a text. SPB assigns higher weights to words that appear in sentences that are either close to the beginning or end of a text. Experiments conducted on six benchmark datasets show that WP and SP do not statistically differ. However, for datasets whose keywords appear early in the text WP performs better than SP with no statistical difference, while for datasets where keywords are evenly distributed in text SP statistically performs better than WP.
format	article
author	Osman KABASAKAL Alev MUTLU
author_facet	Osman KABASAKAL Alev MUTLU
author_sort	Osman KABASAKAL
title	ON THE EFFECT OF WORD POSITIONS IN GRAPH-BASED KEYWORD EXTRACTION
title_short	ON THE EFFECT OF WORD POSITIONS IN GRAPH-BASED KEYWORD EXTRACTION
title_full	ON THE EFFECT OF WORD POSITIONS IN GRAPH-BASED KEYWORD EXTRACTION
title_fullStr	ON THE EFFECT OF WORD POSITIONS IN GRAPH-BASED KEYWORD EXTRACTION
title_full_unstemmed	ON THE EFFECT OF WORD POSITIONS IN GRAPH-BASED KEYWORD EXTRACTION
title_sort	on the effect of word positions in graph-based keyword extraction
publisher	National Defense University Barbaros Naval Sciences and Engineering Institute Journal of Naval Science and Engineering
publishDate	2021
url	https://doaj.org/article/327c54751393413bbd6c9c650b41b4e6
work_keys_str_mv	AT osmankabasakal ontheeffectofwordpositionsingraphbasedkeywordextraction AT alevmutlu ontheeffectofwordpositionsingraphbasedkeywordextraction
_version_	1718426292231077888

ON THE EFFECT OF WORD POSITIONS IN GRAPH-BASED KEYWORD EXTRACTION

Ejemplares similares