Presentation 1997/5/11
Automatic Acquisition of Word Selection Rule by Korean-Japanese bilingual text
Jun-ichi Nakamura, Kyoko Kai, Hiroaki Kagami, Kazuhiro Ishiyama,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this paper, we describe a system for word selection in a Korean-Japanese machine translation system, using bilingual text corpus. This system selects translations of nouns. Kaku-jyoshi's (Korean/Japanese case maker), and verbs as a whole with co-occurrence frequency extracted from a corpus. From 2,219 sentence pairs, 1,561 co-occurrence data are extracted for subject, object, so-case, and ro-case. The extraction is done by word-to-word matching, by use of syntactic similarity of Korean and Japanese. The co-occurrence data is extended by the Japanese thesaurus (Bunrui-Goi-Hyo) dynamically to improve the performance. The evaluation for 31 ambiguous sentences shows that correct words are selected for 58.1% sentences.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Machine Translation / Korean / Word Selection / Statistical Processing / Thesaurus
Paper # NLC97-6
Date of Issue

Conference Information
Committee NLC
Conference Date 1997/5/11(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Natural Language Understanding and Models of Communication (NLC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Automatic Acquisition of Word Selection Rule by Korean-Japanese bilingual text
Sub Title (in English)
Keyword(1) Machine Translation
Keyword(2) Korean
Keyword(3) Word Selection
Keyword(4) Statistical Processing
Keyword(5) Thesaurus
1st Author's Name Jun-ichi Nakamura
1st Author's Affiliation Department of Artificial Intelligence, Kyushu Institute of Technology()
2nd Author's Name Kyoko Kai
2nd Author's Affiliation Information Science Center, Kyushu Institute of Technology
3rd Author's Name Hiroaki Kagami
3rd Author's Affiliation Fukuoka Research Center, Daewoo Electronics Co., Ltd.
4th Author's Name Kazuhiro Ishiyama
4th Author's Affiliation Department of Artificial Intelligence, Kyushu Institute of Technology
Date 1997/5/11
Paper # NLC97-6
Volume (vol) vol.97
Number (no) 50
Page pp.pp.-
#Pages 8
Date of Issue