Unsupervised word sense disambiguation using WordNet relatives

Hee Cheol Seo, Hoojung Chung, Hae-Chang Rim, Sung Hyon Myaeng, Soo Hong Kim

Research output: Contribution to journalArticle

28 Citations (Scopus)

Abstract

This paper describes a sense disambiguation method for a polysemous target noun using the context words surrounding the target noun and its WordNet relatives, such as synonyms, hypernyms and hyponyms. The result of sense disambiguation is a relative that can substitute for that target noun in a context. The selection is made based on co-occurrence frequency between candidate relatives and each word in the context. Since the co-occurrence frequency is obtainable from a raw corpus, the method is considered to be an unsupervised learning algorithm and therefore does not require a sense-tagged corpus. In a series of experiments using SemCor and the corpus of SENSEVAL-2 lexical sample task, all in English, and using some Korean data, the proposed method was shown to be very promising. In particular, its performance was superior to that of the other approaches evaluated on the same test corpora.

Original languageEnglish
Pages (from-to)253-273
Number of pages21
JournalComputer Speech and Language
Volume18
Issue number3 SPEC. ISS.
DOIs
Publication statusPublished - 2004 Jul 1

Fingerprint

Word Sense Disambiguation
Unsupervised learning
WordNet
Learning algorithms
Target
Experiments
Unsupervised Learning
candidacy
Substitute
Learning
Learning Algorithm
experiment
learning
performance
Series
Corpus
Experiment
Context

ASJC Scopus subject areas

  • Signal Processing
  • Electrical and Electronic Engineering
  • Experimental and Cognitive Psychology
  • Linguistics and Language

Cite this

Unsupervised word sense disambiguation using WordNet relatives. / Seo, Hee Cheol; Chung, Hoojung; Rim, Hae-Chang; Myaeng, Sung Hyon; Kim, Soo Hong.

In: Computer Speech and Language, Vol. 18, No. 3 SPEC. ISS., 01.07.2004, p. 253-273.

Research output: Contribution to journalArticle

Seo, Hee Cheol ; Chung, Hoojung ; Rim, Hae-Chang ; Myaeng, Sung Hyon ; Kim, Soo Hong. / Unsupervised word sense disambiguation using WordNet relatives. In: Computer Speech and Language. 2004 ; Vol. 18, No. 3 SPEC. ISS. pp. 253-273.
@article{9fb95489f1bc4f4181af601a0d3fa834,
title = "Unsupervised word sense disambiguation using WordNet relatives",
abstract = "This paper describes a sense disambiguation method for a polysemous target noun using the context words surrounding the target noun and its WordNet relatives, such as synonyms, hypernyms and hyponyms. The result of sense disambiguation is a relative that can substitute for that target noun in a context. The selection is made based on co-occurrence frequency between candidate relatives and each word in the context. Since the co-occurrence frequency is obtainable from a raw corpus, the method is considered to be an unsupervised learning algorithm and therefore does not require a sense-tagged corpus. In a series of experiments using SemCor and the corpus of SENSEVAL-2 lexical sample task, all in English, and using some Korean data, the proposed method was shown to be very promising. In particular, its performance was superior to that of the other approaches evaluated on the same test corpora.",
author = "Seo, {Hee Cheol} and Hoojung Chung and Hae-Chang Rim and Myaeng, {Sung Hyon} and Kim, {Soo Hong}",
year = "2004",
month = "7",
day = "1",
doi = "10.1016/j.csl.2004.05.004",
language = "English",
volume = "18",
pages = "253--273",
journal = "Computer Speech and Language",
issn = "0885-2308",
publisher = "Academic Press Inc.",
number = "3 SPEC. ISS.",

}

TY - JOUR

T1 - Unsupervised word sense disambiguation using WordNet relatives

AU - Seo, Hee Cheol

AU - Chung, Hoojung

AU - Rim, Hae-Chang

AU - Myaeng, Sung Hyon

AU - Kim, Soo Hong

PY - 2004/7/1

Y1 - 2004/7/1

N2 - This paper describes a sense disambiguation method for a polysemous target noun using the context words surrounding the target noun and its WordNet relatives, such as synonyms, hypernyms and hyponyms. The result of sense disambiguation is a relative that can substitute for that target noun in a context. The selection is made based on co-occurrence frequency between candidate relatives and each word in the context. Since the co-occurrence frequency is obtainable from a raw corpus, the method is considered to be an unsupervised learning algorithm and therefore does not require a sense-tagged corpus. In a series of experiments using SemCor and the corpus of SENSEVAL-2 lexical sample task, all in English, and using some Korean data, the proposed method was shown to be very promising. In particular, its performance was superior to that of the other approaches evaluated on the same test corpora.

AB - This paper describes a sense disambiguation method for a polysemous target noun using the context words surrounding the target noun and its WordNet relatives, such as synonyms, hypernyms and hyponyms. The result of sense disambiguation is a relative that can substitute for that target noun in a context. The selection is made based on co-occurrence frequency between candidate relatives and each word in the context. Since the co-occurrence frequency is obtainable from a raw corpus, the method is considered to be an unsupervised learning algorithm and therefore does not require a sense-tagged corpus. In a series of experiments using SemCor and the corpus of SENSEVAL-2 lexical sample task, all in English, and using some Korean data, the proposed method was shown to be very promising. In particular, its performance was superior to that of the other approaches evaluated on the same test corpora.

UR - http://www.scopus.com/inward/record.url?scp=3142656769&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=3142656769&partnerID=8YFLogxK

U2 - 10.1016/j.csl.2004.05.004

DO - 10.1016/j.csl.2004.05.004

M3 - Article

AN - SCOPUS:3142656769

VL - 18

SP - 253

EP - 273

JO - Computer Speech and Language

JF - Computer Speech and Language

SN - 0885-2308

IS - 3 SPEC. ISS.

ER -