Unsupervised Word-Sense Disambiguation Using Bilingual Comparable Corpora

Hiroyuki KAJI; Yasutsugu MORIMOTO

doi:10.1093/ietisy/e88-d.2.289

Unsupervised Word-Sense Disambiguation Using Bilingual Comparable Corpora

Hiroyuki KAJI, Yasutsugu MORIMOTO

Full Text Views

0

Cite this

Summary :

An unsupervised method for word-sense disambiguation using bilingual comparable corpora was developed. First, it extracts word associations, i.e., statistically significant pairs of associated words, from the corpus of each language. Then, it aligns word associations by consulting a bilingual dictionary and calculates correlation between senses of a target polysemous word and its associated words, which can be regarded as clues for identifying the sense of the target word. To overcome the problem of disparity of topical coverage between corpora of the two languages as well as the problem of ambiguity in word-association alignment, an algorithm for iteratively calculating a sense-vs.-clue correlation matrix for each target word was devised. Word-sense disambiguation for each instance of the target word is done by selecting the sense that maximizes the score, i.e., a weighted sum of the correlations between each sense and clues appearing in the context of the instance. An experiment using Wall Street Journal and Nihon Keizai Shimbun corpora together with the EDR bilingual dictionary showed that the new method has promising performance; namely, the F-measure of its sense selection was 74.6% compared to a baseline of 62.8%. The developed method will possibly be extended into a fully unsupervised method that features automatic division and definition of word senses.

Publication: IEICE TRANSACTIONS on Information Vol.E88-D No.2 pp.289-301

Publication Date: 2005/02/01

Publicized

Online ISSN

DOI: 10.1093/ietisy/e88-d.2.289

Type of Manuscript: PAPER

Category: Natural Language Processing

Cite this

Copy

Hiroyuki KAJI, Yasutsugu MORIMOTO, "Unsupervised Word-Sense Disambiguation Using Bilingual Comparable Corpora" in IEICE TRANSACTIONS on Information, vol. E88-D, no. 2, pp. 289-301, February 2005, doi: 10.1093/ietisy/e88-d.2.289.
Abstract: An unsupervised method for word-sense disambiguation using bilingual comparable corpora was developed. First, it extracts word associations, i.e., statistically significant pairs of associated words, from the corpus of each language. Then, it aligns word associations by consulting a bilingual dictionary and calculates correlation between senses of a target polysemous word and its associated words, which can be regarded as clues for identifying the sense of the target word. To overcome the problem of disparity of topical coverage between corpora of the two languages as well as the problem of ambiguity in word-association alignment, an algorithm for iteratively calculating a sense-vs.-clue correlation matrix for each target word was devised. Word-sense disambiguation for each instance of the target word is done by selecting the sense that maximizes the score, i.e., a weighted sum of the correlations between each sense and clues appearing in the context of the instance. An experiment using Wall Street Journal and Nihon Keizai Shimbun corpora together with the EDR bilingual dictionary showed that the new method has promising performance; namely, the F-measure of its sense selection was 74.6% compared to a baseline of 62.8%. The developed method will possibly be extended into a fully unsupervised method that features automatic division and definition of word senses.
URL: https://global.ieice.org/en_transactions/information/10.1093/ietisy/e88-d.2.289/_p

Copy

@ARTICLE{e88-d_2_289,
author={Hiroyuki KAJI, Yasutsugu MORIMOTO, },
journal={IEICE TRANSACTIONS on Information},
title={Unsupervised Word-Sense Disambiguation Using Bilingual Comparable Corpora},
year={2005},
volume={E88-D},
number={2},
pages={289-301},
abstract={An unsupervised method for word-sense disambiguation using bilingual comparable corpora was developed. First, it extracts word associations, i.e., statistically significant pairs of associated words, from the corpus of each language. Then, it aligns word associations by consulting a bilingual dictionary and calculates correlation between senses of a target polysemous word and its associated words, which can be regarded as clues for identifying the sense of the target word. To overcome the problem of disparity of topical coverage between corpora of the two languages as well as the problem of ambiguity in word-association alignment, an algorithm for iteratively calculating a sense-vs.-clue correlation matrix for each target word was devised. Word-sense disambiguation for each instance of the target word is done by selecting the sense that maximizes the score, i.e., a weighted sum of the correlations between each sense and clues appearing in the context of the instance. An experiment using Wall Street Journal and Nihon Keizai Shimbun corpora together with the EDR bilingual dictionary showed that the new method has promising performance; namely, the F-measure of its sense selection was 74.6% compared to a baseline of 62.8%. The developed method will possibly be extended into a fully unsupervised method that features automatic division and definition of word senses.},
keywords={},
doi={10.1093/ietisy/e88-d.2.289},
ISSN={},
month={February},}

Copy

TY - JOUR
TI - Unsupervised Word-Sense Disambiguation Using Bilingual Comparable Corpora
T2 - IEICE TRANSACTIONS on Information
SP - 289
EP - 301
AU - Hiroyuki KAJI
AU - Yasutsugu MORIMOTO
PY - 2005
DO - 10.1093/ietisy/e88-d.2.289
JO - IEICE TRANSACTIONS on Information
SN -
VL - E88-D
IS - 2
JA - IEICE TRANSACTIONS on Information
Y1 - February 2005
AB - An unsupervised method for word-sense disambiguation using bilingual comparable corpora was developed. First, it extracts word associations, i.e., statistically significant pairs of associated words, from the corpus of each language. Then, it aligns word associations by consulting a bilingual dictionary and calculates correlation between senses of a target polysemous word and its associated words, which can be regarded as clues for identifying the sense of the target word. To overcome the problem of disparity of topical coverage between corpora of the two languages as well as the problem of ambiguity in word-association alignment, an algorithm for iteratively calculating a sense-vs.-clue correlation matrix for each target word was devised. Word-sense disambiguation for each instance of the target word is done by selecting the sense that maximizes the score, i.e., a weighted sum of the correlations between each sense and clues appearing in the context of the instance. An experiment using Wall Street Journal and Nihon Keizai Shimbun corpora together with the EDR bilingual dictionary showed that the new method has promising performance; namely, the F-measure of its sense selection was 74.6% compared to a baseline of 62.8%. The developed method will possibly be extended into a fully unsupervised method that features automatic division and definition of word senses.
ER -