Bootstrapping a Semantic Lexicon on Verb Similarities

S. Syed; M. Spruit; M. Borit

doi:10.5220/0006036901890196

Bootstrapping a Semantic Lexicon on Verb Similarities

S. Syed
, M. Spruit
, M. Borit

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

Abstract

We present a bootstrapping algorithm to create a semantic lexicon from a list of seed words and a corpus that was mined from the web. We exploit extraction patterns to bootstrap the lexicon and use collocation statistics to dynamically score new lexicon entries. Extraction patterns are subsequently scored by calculating
the conditional probability in relation to a non-related text corpus. We find that verbs that are highly domain related achieved the highest accuracy and collocation statistics affect the accuracy positively and negatively during the bootstrapping runs.

Original language	English
Title of host publication	Proceedings of the 8th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management
Editors	A. Fred, Jan Dietz, David Aveiro, Kecheng Liu, Jorge Bernardino, Joaquim Filipe
Place of Publication	Porto
Publisher	SciTePress
Pages	189–196
Volume	1
ISBN (Print)	978-989-758-203-5
DOIs	https://doi.org/10.5220/0006036901890196
Publication status	Published - 2016

Keywords

Semantic Lexicon
Bootstrapping
Extraction Patterns
Web Mining

Access to Document

10.5220/0006036901890196

BootstrappingFinal published version, 875 KB

Cite this

Syed, S., Spruit, M., & Borit, M. (2016). Bootstrapping a Semantic Lexicon on Verb Similarities. In A. Fred, J. Dietz, D. Aveiro, K. Liu, J. Bernardino, & J. Filipe (Eds.), Proceedings of the 8th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management (Vol. 1, pp. 189–196). SciTePress. https://doi.org/10.5220/0006036901890196

@inproceedings{d559bc05ebe348e486ba62edd284fd1b,

title = "Bootstrapping a Semantic Lexicon on Verb Similarities",

abstract = "We present a bootstrapping algorithm to create a semantic lexicon from a list of seed words and a corpus that was mined from the web. We exploit extraction patterns to bootstrap the lexicon and use collocation statistics to dynamically score new lexicon entries. Extraction patterns are subsequently scored by calculating the conditional probability in relation to a non-related text corpus. We find that verbs that are highly domain related achieved the highest accuracy and collocation statistics affect the accuracy positively and negatively during the bootstrapping runs.",

keywords = "Semantic Lexicon, Bootstrapping, Extraction Patterns, Web Mining",

author = "S. Syed and M. Spruit and M. Borit",

year = "2016",

doi = "10.5220/0006036901890196",

language = "English",

isbn = "978-989-758-203-5",

volume = "1",

pages = "189–196",

editor = "A. Fred and Jan Dietz and David Aveiro and Kecheng Liu and Jorge Bernardino and Filipe, \{Joaquim \}",

booktitle = "Proceedings of the 8th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management",

publisher = "SciTePress",

}

Bootstrapping a Semantic Lexicon on Verb Similarities. / Syed, S.; Spruit, M.; Borit, M.
Proceedings of the 8th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management. ed. / A. Fred; Jan Dietz; David Aveiro; Kecheng Liu; Jorge Bernardino; Joaquim Filipe. Vol. 1 Porto: SciTePress, 2016. p. 189–196.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Bootstrapping a Semantic Lexicon on Verb Similarities

AU - Syed, S.

AU - Spruit, M.

AU - Borit, M.

PY - 2016

Y1 - 2016

N2 - We present a bootstrapping algorithm to create a semantic lexicon from a list of seed words and a corpus that was mined from the web. We exploit extraction patterns to bootstrap the lexicon and use collocation statistics to dynamically score new lexicon entries. Extraction patterns are subsequently scored by calculating the conditional probability in relation to a non-related text corpus. We find that verbs that are highly domain related achieved the highest accuracy and collocation statistics affect the accuracy positively and negatively during the bootstrapping runs.

AB - We present a bootstrapping algorithm to create a semantic lexicon from a list of seed words and a corpus that was mined from the web. We exploit extraction patterns to bootstrap the lexicon and use collocation statistics to dynamically score new lexicon entries. Extraction patterns are subsequently scored by calculating the conditional probability in relation to a non-related text corpus. We find that verbs that are highly domain related achieved the highest accuracy and collocation statistics affect the accuracy positively and negatively during the bootstrapping runs.

KW - Semantic Lexicon

KW - Bootstrapping

KW - Extraction Patterns

KW - Web Mining

U2 - 10.5220/0006036901890196

DO - 10.5220/0006036901890196

M3 - Conference contribution

SN - 978-989-758-203-5

VL - 1

SP - 189

EP - 196

BT - Proceedings of the 8th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management

A2 - Fred, A.

A2 - Dietz, Jan

A2 - Aveiro, David

A2 - Liu, Kecheng

A2 - Bernardino, Jorge

A2 - Filipe, Joaquim

PB - SciTePress

CY - Porto

ER -

Bootstrapping a Semantic Lexicon on Verb Similarities

Abstract

Keywords

Access to Document

Fingerprint

Cite this