Presentation 2017-02-09
Construction of a Bilingual Term Extension System
Kazuya Ishibashi, Kyo Kageura, Miki Iwai, Koichi Takeuchi,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In most of previous work, pattern-based approaches or statistical learning model based approaches are applied to extracting bilingual terms from documents. There still remain, however, not small terms that are notextracted because of their low frequency in the documents. In contrast to the previous work, we have proposed an approach to extract new bilingual terms from bilingual term dictionaries because most of new terms can be composed of existing concepts, i.e., constituents of terms. One of the key issues of the proposed approach is how to make suitable clusters in bipartite graph of term constituents for generating proper new terms. In this study we applied two methods of clustering, i.e., Kernighan-Lin algorithm and Spectral Co-Clustering to dividing bipartite graph. The experimental results of generating new bilingual terms in five domains show that the Spectral Co-Clustering based system extracts proper new terms with a maximal of 58% accuracy and finds correct their translations with a maximal of 26% accuracy. In the experimental results of new term extraction task of all domains, the Spectral Co-Clustering system outperforms Kernighan-Lin algorithm based system.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Terminorogy / KL algorithm / Spectral Co-Clustering
Paper # NLC2016-42
Date of Issue 2017-02-02 (NLC)

Conference Information
Committee NLC / IPSJ-IFAT
Conference Date 2017/2/9(2days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair Hiroshi Kanayama(IBM)
Vice Chair Makoto Ichise(NTT DoCoMo) / Takeshi Sakaki(Univ. of Tokyo/Hottolink)
Secretary Makoto Ichise(Ryukoku Univ.) / Takeshi Sakaki(Kyushu Inst. of Tech.)
Assistant Ryuichiro Higashinaka(NTT) / Mitsuo Yoshida(Toyohashi Univ. of Tech.)

Paper Information
Registration To Technical Committee on Natural Language Understanding and Models of Communication / Special Interest Group on Information Fundamentals and Access Technologies
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Construction of a Bilingual Term Extension System
Sub Title (in English)
Keyword(1) Terminorogy
Keyword(2) KL algorithm
Keyword(3) Spectral Co-Clustering
1st Author's Name Kazuya Ishibashi
1st Author's Affiliation Okayama University(Okayama Univ.)
2nd Author's Name Kyo Kageura
2nd Author's Affiliation The University of Tokyo(UTokyo)
3rd Author's Name Miki Iwai
3rd Author's Affiliation The University of Tokyo(UTokyo)
4th Author's Name Koichi Takeuchi
4th Author's Affiliation Okayama University(Okayama Univ.)
Date 2017-02-09
Paper # NLC2016-42
Volume (vol) vol.116
Number (no) NLC-451
Page pp.pp.13-17(NLC),
#Pages 5
Date of Issue 2017-02-02 (NLC)