Development of statistical machine translation mode using Bayesian statistics

• Project/Area Number: 19500114
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Single-year Grants
• Section: 一般
• Research Field: Intelligent informatics
• Research Institution: University of Tsukuba
• Principal Investigator: YAMAMOTO Mikio University of Tsukuba, 大学院・システム情報工学研究科, 教授 (40210562)
• Research Collaborator: NORIMATSU Jyunya 筑波大学, 大学院・システム情報工学研究科・コンピュータサイエンス専攻博士前期課程, 学生
• Project Period (FY): 2007 – 2008
• Project Status: Completed (Fiscal Year 2008)
• Budget Amount: ¥4,550,000 (Direct Cost: ¥3,500,000、Indirect Cost: ¥1,050,000); Fiscal Year 2008: ¥1,690,000 (Direct Cost: ¥1,300,000、Indirect Cost: ¥390,000); Fiscal Year 2007: ¥2,860,000 (Direct Cost: ¥2,200,000、Indirect Cost: ¥660,000)
• Keywords: 自然言語処理 / 機械翻訳 / 機械学習 / ベイズ統計学 / 統計的機械翻訳 / 構文 / シンクロナスCFG / 階層フレーズ / CKYパージング / ベイズ学習 / デコーダ

Research Abstract

構文情報を利用した統計的機械翻訳法として代表的な階層フレーズモデルに着目し、日英間のように比較的遠い言語間の翻訳精度を上げる2つの手法を開発した。1つ目は階層フレーズからフレーズ順序知識のみを分離し、より柔軟にフレーズ移動を行えるモデルを開発した。2つ目は単語あるいは翻訳の大きな移動をサポートする特別なルールを含む場合でも効率的にモデル推定を行える階層フレーズモデルのベイズ的推定手法とその利用法を検討/開発した。

Research Products

• [Journal Article] Producing a test collection for patent machine translation in the seventh NTCIR workshop, Proc (2008)
  • Author(s): A. Fujii, M. Utiyama, M. Yamamoto and T. Utsuro
  • Journal Title: of The 6th International Conference on Language Resources and Evaluation Morocco Pages: 671-674
  • Peer Reviewed
• [Journal Article] Toward the Evaluation of Machine Translation Using Patent Information (2008)
  • Author(s): A. Fujii, M. Utiyama, M. Yamamoto and T. Utsuro
  • Journal Title: The 8th Conference of the Association for Machine Translation in the Americas Pages: 97-106
  • Peer Reviewed
• [Presentation] 統計的機械翻訳におけるフレーズ対応最適化を用いた翻訳候補のリランキング (2009)
  • Author(s): 越川満、内山将夫、梅谷俊治、松井知己、山杢幹雄
  • Organizer: 言語処理学会第15回年次大会発表論文集
  • Place of Presentation: 鳥取
  • Year and Date: 2009-03-05
• [Presentation] 大規模な共通基盤による機械翻訳システムの比較評価 : NTCIR特許翻訳タスク最新事情 (2009)
  • Author(s): 藤井敦、内山将夫、山杢髄、宇津呂武仁
  • Organizer: 言語処理学会第15回年次大会発表論文集
  • Place of Presentation: 鳥取
  • Year and Date: 2009-03-03
• [Presentation] Overview of the Patent Translation Task at the NTCIR-7 Workshop (2008)
  • Author(s): A. Fujii, M. Utiyama, M. Yamamoto and T. Utsuro
  • Organizer: Proceedings of NTCIR-7 Workshop Meeting
  • Place of Presentation: Tokyo
  • Year and Date: 2008-12-18
• [Presentation] Diversion of hierarchical phrases as reordering template (2008)
  • Author(s): M. Yamamoto et.al.
  • Organizer: Proceedings of NTCIR-7 Workshop Meeting
  • Place of Presentation: Tokyo
  • Year and Date: 2008-12-18
• [Presentation] Diversion of hierarchical phrases as reordering templates (2008)
  • Author(s): Mikio Yamamoto
  • Organizer: Proceedings of NTCIR-7 Workshop Meeting
  • Place of Presentation: Tokyo
  • Year and Date: 2008-12-18
• [Presentation] Toward the Evaluation of Machine Translation Using Patent Information (2008)
  • Author(s): Atsushi Fujii,
  • Organizer: The 8th Conference of the Association for Machine Translation in the Americas
  • Place of Presentation: Hawaii, USA.
  • Year and Date: 2008-10-27
• [Presentation] Integer program-ming for a phrase alignment problem on statistical machine translation (2008)
  • Author(s): M. Yamamoto
  • Organizer: Mathematical Programming in the 21st Century : Optimizatioin Modeling and Algorithms
  • Place of Presentation: Kyoto
  • Year and Date: 2008-07-24
• [Presentation] Integer programming for a phrase alignment problem on statistical machine translation (2008)
  • Author(s): Mikio Yamamoto
  • Organizer: Mathematical Programming in the 21st Century : Optimizatioin Modeling and Algorithms
  • Place of Presentation: kyoto
  • Year and Date: 2008-07-22
• [Presentation] Sentiment analysis based on probabilistic models using inter-sentence information (2008)
  • Author(s): Kugatsu Sadamitsu
  • Organizer: The 6th International Conference on Language Resources and Evaluation
  • Place of Presentation: Morocco
  • Year and Date: 2008-05-30
• [Presentation] Producing a test collection for patent machine translation in the seventh NTCIR workshop (2008)
  • Author(s): Atsushi Fujii
  • Organizer: The 6th International Conference on Language Resources and Evaluation
  • Place of Presentation: Morocco
  • Year and Date: 2008-05-29
• [Presentation] 特許情報を対象とした機械翻訳:共通基盤による評価タスクを目指して (2007)
  • Author(s): 内山 将夫
  • Organizer: 情報処理学会自然言語処理研究会
  • Place of Presentation: 徳島大学工学部(徳島市)
  • Year and Date: 2007-07-25