Title

Comparing and Extracting Paraphrasing Words with 2-Way Bilingual Dictionaries

Authors

Kazutaka TAKAO (ATR Spoken Language Translation Research Laboratories, 2-2-2 Hikaridai, Seika-cho, Soraku-gun, Kyoto, 619-0288, JAPAN)

 Kenji IMAMURA (ATR Spoken Language Translation Research Laboratories, 2-2-2 Hikaridai, Seika-cho, Soraku-gun, Kyoto, 619-0288, JAPAN)

Hideki KASHIOKA (ATR Spoken Language Translation Research Laboratories, 2-2-2 Hikaridai, Seika-cho, Soraku-gun, Kyoto, 619-0288, JAPAN)

Session

WP2: Lexicons

Abstract

We analyze a variety of lexical expressions with 2-way bilingual dictionaries and propose a method for extracting paraphrasing words. First, we compare the coverage between an English-Japanese dictionary and a Japanese-English dictionary from the viewpoint of the returnability of the words by translating English to Japanese, and then back to English again. The variety is shown using examples. Next, we propose a method of automatically extracting English paraphrasing word groups; we gathered the English index words which have the same Japanese translation words in the E-J dictionary. The English words which are difficult to distinguish for native speakers of Japanese were then extracted into a paraphrasing group. We also extract the Japanese paraphrasing word groups for comparison. This method will be useful for sentence matching, especially in order to accept the variety of expressions.

Keywords

Two-Way bilingual dictionaries, Paraphrasing words, Japanese-English dictionary, English-Japanese dictionary, Machine translation

Full Paper

59.pdf