Presentation 2001/10/10
The Successes and Failures of Lexical and Stractural Translation Retrieval
Timothy BALDWIN, Atsushi OKAZAKI, Takenobu TOKUNAGA, Hozumi TANAKA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper describes two distinct translation retrieval methods within the context of the Senseval2 Japanese translation task. The first is based on lexical similarity and models strings as a bag of character bigrams, whereas the second is based on structural similarity and determines similarity via parse trees and conceptual similarity. We also discuss a hybrid approach, combining the results of the lexical and structural methods. Despite its simplistic nature, the lexical method was found to outperform the structural method by a clear margin, but the hybrid method to be the strongest overall contender.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) translation retrieval / lexical similarity / structural similarity
Paper # NLC 2001-38
Date of Issue

Conference Information
Committee NLC
Conference Date 2001/10/10(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 ENG
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) The Successes and Failures of Lexical and Stractural Translation Retrieval
Sub Title (in English)
Keyword(1) translation retrieval
Keyword(2) lexical similarity
Keyword(3) structural similarity
1st Author's Name Timothy BALDWIN
1st Author's Affiliation Centre for the Study of Languageand Information (CSLI)()
2nd Author's Name Atsushi OKAZAKI
2nd Author's Affiliation Department of Computer Science, Graduate School of Information Science and Engineering
3rd Author's Name Takenobu TOKUNAGA
3rd Author's Affiliation Department of Computer Science, Graduate School of Information Science and Engineering
4th Author's Name Hozumi TANAKA
4th Author's Affiliation Department of Computer Science, Graduate School of Information Science and Engineering
Date 2001/10/10
Paper # NLC 2001-38
Volume (vol) vol.101
Number (no) 351
Page pp.pp.-
#Pages 8
Date of Issue