Translation divergence analysis and processing for Mandarin-English parallel text exploitation

Shun Chieh Lin; Jia Ching Wang; Jhing Fa Wang

Translation divergence analysis and processing for Mandarin-English parallel text exploitation

Shun Chieh Lin, Jia Ching Wang, Jhing Fa Wang

Department of Computer Science and Information Engineering

Research output: Contribution to conference › Paper › peer-review

1 Scopus citations

Abstract

Previous work shows that the process of parallel text exploitation to extract transfer mappings between language pairs raises the capability of language translation. However, while this process can be fully automated, one thorny problem called "divergence" causes indisposed mapping extraction. Therefore, this paper discuss the issues of parallel text exploitation, in general, with special emphasis on divergence analysis and processing. In the experiments on a Mandarin-English travel conversation corpus of 11,885 sentence pairs, the perplexity with the alignments in IBM translation model is reduced averagely from 13.65 to 5.18 by sieving out inappropriate sentences from the collected corpus.

Original language	English
State	Published - 2005
Event	17th Conference on Computational Linguistics and Speech Processing, ROCLING 2005 - Tainan, Taiwan Duration: 15 Sep 2005 → 16 Sep 2005

Conference

Conference	17th Conference on Computational Linguistics and Speech Processing, ROCLING 2005
Country/Territory	Taiwan
City	Tainan
Period	15/09/05 → 16/09/05

Cite this

@conference{9337d0cdcf6b45a6b70e294b4a0b6e79,

title = "Translation divergence analysis and processing for Mandarin-English parallel text exploitation",

abstract = "Previous work shows that the process of parallel text exploitation to extract transfer mappings between language pairs raises the capability of language translation. However, while this process can be fully automated, one thorny problem called {"}divergence{"} causes indisposed mapping extraction. Therefore, this paper discuss the issues of parallel text exploitation, in general, with special emphasis on divergence analysis and processing. In the experiments on a Mandarin-English travel conversation corpus of 11,885 sentence pairs, the perplexity with the alignments in IBM translation model is reduced averagely from 13.65 to 5.18 by sieving out inappropriate sentences from the collected corpus.",

author = "Lin, {Shun Chieh} and Wang, {Jia Ching} and Wang, {Jhing Fa}",

year = "2005",

language = "???core.languages.en_GB???",

note = "null ; Conference date: 15-09-2005 Through 16-09-2005",

}

TY - CONF

T1 - Translation divergence analysis and processing for Mandarin-English parallel text exploitation

AU - Lin, Shun Chieh

AU - Wang, Jia Ching

AU - Wang, Jhing Fa

PY - 2005

Y1 - 2005

N2 - Previous work shows that the process of parallel text exploitation to extract transfer mappings between language pairs raises the capability of language translation. However, while this process can be fully automated, one thorny problem called "divergence" causes indisposed mapping extraction. Therefore, this paper discuss the issues of parallel text exploitation, in general, with special emphasis on divergence analysis and processing. In the experiments on a Mandarin-English travel conversation corpus of 11,885 sentence pairs, the perplexity with the alignments in IBM translation model is reduced averagely from 13.65 to 5.18 by sieving out inappropriate sentences from the collected corpus.

AB - Previous work shows that the process of parallel text exploitation to extract transfer mappings between language pairs raises the capability of language translation. However, while this process can be fully automated, one thorny problem called "divergence" causes indisposed mapping extraction. Therefore, this paper discuss the issues of parallel text exploitation, in general, with special emphasis on divergence analysis and processing. In the experiments on a Mandarin-English travel conversation corpus of 11,885 sentence pairs, the perplexity with the alignments in IBM translation model is reduced averagely from 13.65 to 5.18 by sieving out inappropriate sentences from the collected corpus.

UR - http://www.scopus.com/inward/record.url?scp=84883015610&partnerID=8YFLogxK

M3 - 會議論文

AN - SCOPUS:84883015610

Y2 - 15 September 2005 through 16 September 2005

ER -

Translation divergence analysis and processing for Mandarin-English parallel text exploitation

Abstract

Conference

Fingerprint

Cite this