• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to previous page

Analysis of human behaviour in language understanding and its application for natural language processing

Research Project

Project/Area Number 16H02865
Research Category

Grant-in-Aid for Scientific Research (B)

Allocation TypeSingle-year Grants
Section一般
Research Field Intelligent informatics
Research InstitutionTokyo Institute of Technology

Principal Investigator

Tokunaga Takenobu  東京工業大学, 情報理工学院, 教授 (20197875)

Co-Investigator(Kenkyū-buntansha) 西川 仁  東京工業大学, 情報理工学院, 助教 (00765026)
相澤 彰子  国立情報学研究所, コンテンツ科学研究系, 教授 (90222447)
Project Period (FY) 2016-04-01 – 2019-03-31
Project Status Completed (Fiscal Year 2018)
Budget Amount *help
¥13,260,000 (Direct Cost: ¥10,200,000、Indirect Cost: ¥3,060,000)
Fiscal Year 2018: ¥2,210,000 (Direct Cost: ¥1,700,000、Indirect Cost: ¥510,000)
Fiscal Year 2017: ¥2,470,000 (Direct Cost: ¥1,900,000、Indirect Cost: ¥570,000)
Fiscal Year 2016: ¥8,580,000 (Direct Cost: ¥6,600,000、Indirect Cost: ¥1,980,000)
Keywords自然言語処理 / コーパス / アノテーション / 視線情報 / 振舞い情報 / 機械学習 / 行動履歴 / 視線計測 / コーパスアノテーション
Outline of Final Research Achievements

The main stream of natural language processing (NLP) research is the machine learning-based approach in which a model for a specific task is trained by using manually annotated corpora. To cover broad types of NLP tasks, we classify the task into three: segmentation, linking and transforming. For each kind of task, we collected annotator's eye gaze, key inputs and mouse operations during their annotation activity. We proposed to utilise such annotator's behaviour information for machine learning and verified its effectiveness through the experiments.

Academic Significance and Societal Importance of the Research Achievements

文章読解や問題解決における視線情報などの振舞いを分析する研究はおこなわれているが,アノテーション中の振舞いを記録し,さらにそれを言語処理に活用する研究は類を見ない.本研究課題の成果であるアノテーション過程の作業者の振舞いから言語処理に有用な情報を抽出する手法は,十分な解析性能が得られていなかった言語処理の分野においても解析性能を向上させることができ,その学術的貢献は大きい.自然言語処理技術は実社会ですでに利用され始めている.今後,意味や文脈に踏み込んだ研究が発展することにより,既存の実用システムがさらに高度化され,使い易くなることが期待できる.

Report

(4 results)
  • 2018 Annual Research Report   Final Research Report ( PDF )
  • 2017 Annual Research Report
  • 2016 Annual Research Report
  • Research Products

    (7 results)

All 2019 2017 2016 Other

All Int'l Joint Research (1 results) Presentation (5 results) (of which Int'l Joint Research: 2 results) Book (1 results)

  • [Int'l Joint Research] University of Cambridge(英国)

    • Related Report
      2018 Annual Research Report
  • [Presentation] An argument annotation scheme for the repair of student essays by sentence reordering2019

    • Author(s)
      Jan Wira Gotama Putra, Simone Teufel and Tokunaga Takenobu
    • Organizer
      言語処理学会第25回年次大会(NLP2019)
    • Related Report
      2018 Annual Research Report
  • [Presentation] 視線情報を用いた述語項構造解析モデルへの単語分散表現の導入2017

    • Author(s)
      牧諒亮, 西川仁, 徳永健伸
    • Organizer
      言語処理学会 第23回年次大会
    • Place of Presentation
      筑波大学
    • Year and Date
      2017-03-14
    • Related Report
      2016 Annual Research Report
  • [Presentation] An eye-tracking study of named entity annotation2017

    • Author(s)
      Takenobu Tokunaga, Hitoshi Nishikawa and Tomoya Iwakura
    • Organizer
      Proceedings of Recent Advances in Natural Language Processing (RANLP 2017)
    • Related Report
      2017 Annual Research Report
    • Int'l Joint Research
  • [Presentation] 視線情報を用いた日本語述語項構造解析モデルのパラメータ推定2016

    • Author(s)
      牧 諒亮,西川仁,徳永健伸
    • Organizer
      情報処理学会自然言語処理研究会
    • Place of Presentation
      NTT武蔵野研究センタ
    • Year and Date
      2016-12-21
    • Related Report
      2016 Annual Research Report
  • [Presentation] Parameter estimation of Japanese predicate argument structure analysis model using eye gaze information2016

    • Author(s)
      Ryosuke Maki, Hitoshi Nishikawa and Takenobu Tokunaga
    • Organizer
      the 26th International Conference on Computational Linguistics (COLING 2016)
    • Place of Presentation
      大阪国際会議場
    • Year and Date
      2016-12-13
    • Related Report
      2016 Annual Research Report
    • Int'l Joint Research
  • [Book] Ongoing efforts: Toward behaviour-based corpus evaluation (Handbook of Linguistic Annotation)2017

    • Author(s)
      Takenobu Tokunaga (Nancy Ide and James Pustejovsky Ed.)
    • Total Pages
      1459
    • Publisher
      Springer
    • ISBN
      9789402408799
    • Related Report
      2017 Annual Research Report

URL: 

Published: 2016-04-21   Modified: 2020-03-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi