Improving the distribution of N-grams in phrase tables obtained by the sampling-based method

Juan Luo*, Adrien Lardilleux, Yves Lepage

*この研究の対応する著者

研究成果

抄録

We describe an approach to improve the performance of sampling-based sub-sentential alignment method on translation tasks by investigating the distribution of n-grams in the phrase tables. This approach consists in enforcing the alignment of n-grams. We compare the quality of phrase translation tables output by this approach and that of the state-of-the-art estimation approach in statistical machine translation tasks. We report significant improvements for this approach and show that merging phrase tables outperforms the state-of-the-art techniques.

本文言語English
ホスト出版物のタイトルHuman Language Technology Challenges for Computer Science and Linguistics - 5th Language and Technology Conference, LTC 2011, Revised Selected Papers
出版社Springer Verlag
ページ419-431
ページ数13
ISBN(印刷版)9783319089577
DOI
出版ステータスPublished - 2014
イベント5th Language and Technology Conference, LTC 2011 - Poznan, Poland
継続期間: 2011 11 252011 11 27

出版物シリーズ

名前Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
8387 LNAI
ISSN(印刷版)0302-9743
ISSN(電子版)1611-3349

Conference

Conference5th Language and Technology Conference, LTC 2011
国/地域Poland
CityPoznan
Period11/11/2511/11/27

ASJC Scopus subject areas

  • 理論的コンピュータサイエンス
  • コンピュータ サイエンス(全般)

フィンガープリント

「Improving the distribution of N-grams in phrase tables obtained by the sampling-based method」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル