Improvement of Linguistic Analysis Based on Automatically Acquired Knowledge from Large Text

2012 Fiscal Year Final Research Report

• Project/Area Number: 21700163
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Single-year Grants
• Research Field: Intelligent informatics
• Research Institution: Kyoto University
• Principal Investigator: SHIBATA Tomohide 京都大学, 大学院・情報学研究科, 助教 (70452315)
• Project Period (FY): 2009 – 2012
• Keywords: 自然言語処理 / 大規模テキスト / 知識獲得 / 同義語 / 分布類似度

Research Abstract

This research aims to acquire linguistic knowledge from a large text, and utilize it for improving the accuracy of linguistic analysis. The linguistic knowledge acquired from a large Web text and Wikipedia includes the similarity between words, knowledge between events, and a large-scale lexical knowledge such as synonym and hypernym, and it is managed in integrated and consistent manner. Furthermore, the acquired linguistic knowledge was utilized for improving the accuracy of linguistic analysis such as textual entailment recognition and named entity recognition.

Research Products

• [Journal Article] Predicate-argument Structure based Textual Entailment Recognition System exploiting Wide-coverage Lexical Knowledge (2012)
  • Author(s): Tomohide Shibata and Sadao Kurohashi
  • Journal Title: Special Issue of ACM TALIP on RITE (Recognizing Inference in TExt) Volume: Vol.11, No.4 Pages: 16:1-16:23
  • Peer Reviewed
• [Journal Article] 同一文抽出に基づく類似ページの検出と分類 (2010)
  • Author(s): 柴田知秀, 姜ナウン, 黒橋禎夫
  • Journal Title: 人工知能学会論文誌 Volume: Vol.25, No.1 Pages: 224-232
  • Peer Reviewed
• [Journal Article] 二段階の機械学習を用いたボトムアップ型の固有表現認識 (2009)
  • Author(s): 3船山弘孝, 柴田知秀, 黒橋禎夫
  • Journal Title: 第8回情報科学技術フォーラム (FIT2009) Volume: 第2分冊 Pages: 19-26
  • Peer Reviewed
• [Presentation] 日本語語彙知識の統一的・整合的管理のデザイン (2013)
  • Author(s): 黒橋禎夫, 進義治, 柴田知秀, 村脇有吾, 河原大輔
  • Organizer: 言語処理学会 第19回年次大会
  • Year and Date: 20130300
• [Presentation] 実テキスト解析をささえる語彙知識の自動獲得 (2012)
  • Author(s): 柴田知秀, 村脇有吾, 黒橋禎夫, 河原大輔
  • Organizer: 言語処理学会 第18回年次大会
  • Year and Date: 20120300
• [Presentation] Predicate-argument Structure based Textual Entailment Recognition System of KYOTO Team for NTCIR9 RITE (2011)
  • Author(s): Tomohide Shibata and Sadao Kurohashi
  • Organizer: In Proceedings of the 9th NTCIR Workshop Meeting on Evaluation of Information Access Technologies (NTCIR-9)
  • Place of Presentation: Tokyo, Japan
  • Year and Date: 20111200
• [Presentation] Acquiring Strongly-related Events using Predicate-argument Co-occurring Statistics and Case Frames (2011)
  • Author(s): Tomohide Shibata and Sadao Kurohashi
  • Organizer: In Proceedings of the 5th International Joint Conference on Natural Language Processing (IJCNLP2011, poster)
  • Place of Presentation: Chiang Mai, Thailand
  • Year and Date: 20111100
• [Presentation] 文脈に依存した述語の同義関係獲得 (2010)
  • Author(s): 柴田知秀, 黒橋禎夫
  • Organizer: 情報処理学会 自然言語処理研究会
  • Year and Date: 20101100
• [Presentation] Bottom-up Named Entity Recognition using a Two-stage Machine Learning Method (2009)
  • Author(s): Hirotaka Funayama, Tomohide Shibata, and Sadao Kurohashi
  • Organizer: In Proceedings of Association for Computational Linguistics/International Joint Conference on Natural Language Processing(ACL/IJCNLP2009): Workshop on Multiword Expressions
  • Place of Presentation: Singapore
  • Year and Date: 20090800
• [Remarks]
  • URL: http://nlp.ist.i.kyoto-u.ac.jp/member/shibata/index-j.html
• [Remarks]
  • URL: http://nlp.ist.i.kyoto-u.ac.jp/member/shibata/index.html