Graph-based word clustering using a web search engine

Yutaka Matsuo; Takeshi Sakaki; Kôki Uchiyama; Mitsuru Ishizuka

Graph-based word clustering using a web search engine

Yutaka Matsuo^*, Takeshi Sakaki, Kôki Uchiyama, Mitsuru Ishizuka

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

78 Citations (Scopus)

Abstract

Word clustering is important for automatic thesaurus construction, text classification, and word sense disambiguation. Recently, several studies have reported using the web as a corpus. This paper proposes an unsupervised algorithm for word clustering based on a word similarity measure by web counts. Each pair of words is queried to a search engine, which produces a co-occurrence matrix. By calculating the similarity of words, a word cooccurrence graph is obtained. A new kind of graph clustering algorithm called Newman clustering is applied for efficiently identifying word clusters. Evaluations are made on two sets of word groups derived from a web directory and WordNet.

Original language	English
Title of host publication	COLING/ACL 2006 - EMNLP 2006: 2006 Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference
Pages	542-550
Number of pages	9
Publication status	Published - 2006
Externally published	Yes
Event	11th Conference on Empirical Methods in Natural Language Proceessing, EMNLP 2006, Held in Conjunction with COLING/ACL 2006 - Sydney, NSW Duration: 2006 Jul 22 → 2006 Jul 23

Other

Other	11th Conference on Empirical Methods in Natural Language Proceessing, EMNLP 2006, Held in Conjunction with COLING/ACL 2006
City	Sydney, NSW
Period	06/7/22 → 06/7/23

ASJC Scopus subject areas

Computational Theory and Mathematics
Computer Science Applications
Information Systems

Cite this

Matsuo, Y, Sakaki, T, Uchiyama, K & Ishizuka, M 2006, Graph-based word clustering using a web search engine. in COLING/ACL 2006 - EMNLP 2006: 2006 Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference. pp. 542-550, 11th Conference on Empirical Methods in Natural Language Proceessing, EMNLP 2006, Held in Conjunction with COLING/ACL 2006, Sydney, NSW, 06/7/22.

@inproceedings{6224b64bbe76470d8e7480dbd8131633,

title = "Graph-based word clustering using a web search engine",

abstract = "Word clustering is important for automatic thesaurus construction, text classification, and word sense disambiguation. Recently, several studies have reported using the web as a corpus. This paper proposes an unsupervised algorithm for word clustering based on a word similarity measure by web counts. Each pair of words is queried to a search engine, which produces a co-occurrence matrix. By calculating the similarity of words, a word cooccurrence graph is obtained. A new kind of graph clustering algorithm called Newman clustering is applied for efficiently identifying word clusters. Evaluations are made on two sets of word groups derived from a web directory and WordNet.",

author = "Yutaka Matsuo and Takeshi Sakaki and K{\^o}ki Uchiyama and Mitsuru Ishizuka",

year = "2006",

language = "English",

isbn = "1932432736",

pages = "542--550",

booktitle = "COLING/ACL 2006 - EMNLP 2006: 2006 Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference",

note = "11th Conference on Empirical Methods in Natural Language Proceessing, EMNLP 2006, Held in Conjunction with COLING/ACL 2006 ; Conference date: 22-07-2006 Through 23-07-2006",

}

TY - GEN

T1 - Graph-based word clustering using a web search engine

AU - Matsuo, Yutaka

AU - Sakaki, Takeshi

AU - Uchiyama, Kôki

AU - Ishizuka, Mitsuru

PY - 2006

Y1 - 2006

N2 - Word clustering is important for automatic thesaurus construction, text classification, and word sense disambiguation. Recently, several studies have reported using the web as a corpus. This paper proposes an unsupervised algorithm for word clustering based on a word similarity measure by web counts. Each pair of words is queried to a search engine, which produces a co-occurrence matrix. By calculating the similarity of words, a word cooccurrence graph is obtained. A new kind of graph clustering algorithm called Newman clustering is applied for efficiently identifying word clusters. Evaluations are made on two sets of word groups derived from a web directory and WordNet.

AB - Word clustering is important for automatic thesaurus construction, text classification, and word sense disambiguation. Recently, several studies have reported using the web as a corpus. This paper proposes an unsupervised algorithm for word clustering based on a word similarity measure by web counts. Each pair of words is queried to a search engine, which produces a co-occurrence matrix. By calculating the similarity of words, a word cooccurrence graph is obtained. A new kind of graph clustering algorithm called Newman clustering is applied for efficiently identifying word clusters. Evaluations are made on two sets of word groups derived from a web directory and WordNet.

UR - http://www.scopus.com/inward/record.url?scp=80053343581&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80053343581&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:80053343581

SN - 1932432736

SN - 9781932432732

SP - 542

EP - 550

BT - COLING/ACL 2006 - EMNLP 2006: 2006 Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference

T2 - 11th Conference on Empirical Methods in Natural Language Proceessing, EMNLP 2006, Held in Conjunction with COLING/ACL 2006

Y2 - 22 July 2006 through 23 July 2006

ER -

Graph-based word clustering using a web search engine

Abstract

Other

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this