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

Text Summarization Based on the Combination of Neural Models and Optimization Technologies

Research Project

Project/Area Number 17H01786
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

Okumura Manabu  東京工業大学, 科学技術創成研究院, 教授 (60214079)

Co-Investigator(Kenkyū-buntansha) 平尾 努  日本電信電話株式会社NTTコミュニケーション科学基礎研究所, 協創情報研究部, 主任研究員 (40396148)
高村 大也  東京工業大学, 科学技術創成研究院, 教授 (80361773)
Project Period (FY) 2017-04-01 – 2020-03-31
Project Status Completed (Fiscal Year 2019)
Budget Amount *help
¥17,550,000 (Direct Cost: ¥13,500,000、Indirect Cost: ¥4,050,000)
Fiscal Year 2019: ¥5,070,000 (Direct Cost: ¥3,900,000、Indirect Cost: ¥1,170,000)
Fiscal Year 2018: ¥4,810,000 (Direct Cost: ¥3,700,000、Indirect Cost: ¥1,110,000)
Fiscal Year 2017: ¥7,670,000 (Direct Cost: ¥5,900,000、Indirect Cost: ¥1,770,000)
Keywords自然言語処理 / テキスト要約 / 深層学習 / 離散最適化
Outline of Final Research Achievements

In order to improve the performance of summarization methods using a neural model, 1) we proposed neural summarization models incorporating syntactic and discourse information. 2) We proposed a new method for discourse structure analysis to construct the model for 1). In 1), we proposed a sentence compression method that considers the information of syntactic trees and a sentence selection method that considers the information of discourse structure trees. In 2), we proposed two
discourse structure analysis methods, with and without supervision.
The proposed supervised method achieves the current state-of-the-art performance.

Academic Significance and Societal Importance of the Research Achievements

Sequence-to-Sequence (Seq2Seq)モデルに基づく文圧縮では,すでに圧縮文に採用された単語列とこれから圧縮文に採用しようとする単語との間の文法的な依存関係を明示的に捉える事が難しい為,デコード時に階層的な注意機構に基づき構文的な先読みを行う事が可能なモデルを提案した.Seq2Seqモデルに基づく抽出型手法が単一文書要約において良い性能を示しているが,文間の談話構造は明示的には利用しない.談話構造に関する情報の欠如は,重要度スコア決定における性能劣化や出力要約の一貫性の低下を引き起こす為,原文書の談話構造と文の重要度スコアリング器を同時に学習する新たな枠組みを提案した.

Report

(4 results)
  • 2019 Annual Research Report   Final Research Report ( PDF )
  • 2018 Annual Research Report
  • 2017 Annual Research Report
  • Research Products

    (22 results)

All 2020 2019 2018 2017

All Journal Article (8 results) (of which Peer Reviewed: 8 results,  Open Access: 2 results) Presentation (12 results) (of which Int'l Joint Research: 4 results) Patent(Industrial Property Rights) (2 results)

  • [Journal Article] Syntactically Look-Ahead Attention Network for Sentence Compression2020

    • Author(s)
      Hidetaka Kamigaito and Manabu Okumura
    • Journal Title

      Proc. of the Thirty-Fourth AAAI Conference on Artificial Intelligence (AAAI-20)

      Volume: -

    • Related Report
      2019 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Top-down RST Parsing Utilizing Granularity Levels in Documents2020

    • Author(s)
      Naoki Kobayashi, Tsutomu Hirao, Hidetaka Kamigaito, Manabu Okumura and Masaaki Nagata
    • Journal Title

      Proc. of the Thirty-Fourth AAAI Conference on Artificial Intelligence (AAAI-20)

      Volume: -

    • Related Report
      2019 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Split or Merge: Which is Better for Unsupervised RST Parsing?2019

    • Author(s)
      Naoki Kobayashi, Tsutomu Hirao, Kengo Nakamura, Hidetaka Kamigaito, Manabu Okumura and Masaaki Nagata
    • Journal Title

      Proc. of EMNLP-IJCNLP 2019, 2019

      Volume: - Pages: 5797-5802

    • DOI

      10.18653/v1/d19-1587

    • Related Report
      2019 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Discourse-aware Hierarchical Attention Network for Extractive Single-Document Summarization2019

    • Author(s)
      Tatsuya Ishigaki, Hidetaka Kamigaito, Hiroya Takamura and Manabu Okumura
    • Journal Title

      Proc. of RANLP 2019, 2019

      Volume: - Pages: 497-506

    • DOI

      10.26615/978-954-452-056-4_059

    • Related Report
      2019 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] テキストセグメンテーションによる教師なし修辞構造 解析2019

    • Author(s)
      小林尚輝, 平尾努, 中村健吾, 上垣外英剛, 奥村学, 永田昌明
    • Journal Title

      言語処理学会第25回年次大会発表論文集

      Volume: - Pages: 998-1001

    • Related Report
      2018 Annual Research Report
    • Peer Reviewed
  • [Journal Article] 階層構造を考慮したトップダウン談話構造解析2019

    • Author(s)
      小林尚輝, 平尾努, 上垣外英剛, 奥村学, 永田昌明
    • Journal Title

      言語処理学会第25回年次大会発表論文集

      Volume: - Pages: 1002-1005

    • Related Report
      2018 Annual Research Report
    • Peer Reviewed
  • [Journal Article] 出力長制御を考慮した見出し生成モデルのための大規模コーパス2019

    • Author(s)
      人見雄太, 田口雄哉, 田森秀明, 菊田洸, 西鳥羽二郎, 岡崎直観, 乾健太郎, 奥村学
    • Journal Title

      言語処理学会第25回年次大会発表論文集

      Volume: - Pages: 6-11

    • Related Report
      2018 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Automatic Pyramid Evaluation Exploiting EDU-based Extractive Reference Summaries2018

    • Author(s)
      Hirao, T., Kamigaito, H. and Nagata, M.
    • Journal Title

      Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing

      Volume: - Pages: 4177-4186

    • Related Report
      2018 Annual Research Report
    • Peer Reviewed
  • [Presentation] 階層的な注意機構に基づき統語的な先読みを行う文圧縮手法2020

    • Author(s)
      上垣外英剛,奥村学
    • Organizer
      第243回自然言語処理研究会
    • Related Report
      2019 Annual Research Report
  • [Presentation] 生成型文要約のための抽出性に着目したデータ選択2019

    • Author(s)
      長谷川駿, 上垣外英剛, 奥村学
    • Organizer
      情報処理学会 第241回自然言語処理研究会
    • Related Report
      2019 Annual Research Report
  • [Presentation] 談話構造を考慮する階層的注意機構による抽出型ニューラル単一文書要約2019

    • Author(s)
      石垣達也, 上垣外英剛, 高村大也, 奥村学
    • Organizer
      言語処理学会第25回年次大会
    • Related Report
      2018 Annual Research Report
  • [Presentation] Higher-order Syntactic Attention Network for Longer Sentence Compression2018

    • Author(s)
      Kamigaito, H., Hayashi, K., Hirao, T. and Nishino, M.
    • Organizer
      the 16th Annual Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (to appear).
    • Related Report
      2017 Annual Research Report
    • Int'l Joint Research
  • [Presentation] 依存構造の連鎖を考慮したニューラル文圧縮2018

    • Author(s)
      上垣外英剛, 林克彦, 平尾努, 永田昌明
    • Organizer
      言語処理学会第24回年次大会
    • Related Report
      2017 Annual Research Report
  • [Presentation] 抽出型オラクルを利用した要約の自動評価2018

    • Author(s)
      平尾努, 上垣外英剛, 永田昌明
    • Organizer
      言語処理学会第24回年次大会
    • Related Report
      2017 Annual Research Report
  • [Presentation] 文書要約のための一貫性モデル2018

    • Author(s)
      金澤尚史, 高村大也, 奥村学
    • Organizer
      言語処理学会第24回年次大会
    • Related Report
      2017 Annual Research Report
  • [Presentation] Minimum Risk Training に基づく要約モデルの出力長制御2018

    • Author(s)
      牧野拓哉, 岩倉友哉, 高村大也, 奥村学
    • Organizer
      言語処理学会第24回年次大会
    • Related Report
      2017 Annual Research Report
  • [Presentation] Supervised Attention for Sequence-to-sequence Constituency Parsing2017

    • Author(s)
      Hidetaka, K., Hayashi, K., Hirao, T., Takamura, H., Okumura, M. and Nagata, M.
    • Organizer
      the 8th International Joint Conference on Natural Language Processing (IJCNLP)
    • Related Report
      2017 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Summarizing Lengthy Questions2017

    • Author(s)
      Tatsuya Ishigaki, Hiroya Takamura and Manabu Okumura
    • Organizer
      the 8th International Joint Conference on Natural Language Processing (IJCNLP)
    • Related Report
      2017 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Japanese Sentence Compression with a Large Training Dataset2017

    • Author(s)
      Shun Hasegawa, Yuta Kikuchi, Hiroya Takamura and Manabu Okumura
    • Organizer
      ACL 2017
    • Related Report
      2017 Annual Research Report
    • Int'l Joint Research
  • [Presentation] 「長文質問」のための抽出型及び生成型要約2017

    • Author(s)
      石垣 達也,高村 大也,奥村 学
    • Organizer
      情報処理学会第232回自然言語処理研究会
    • Related Report
      2017 Annual Research Report
  • [Patent(Industrial Property Rights)] 談話構造解析装置2019

    • Inventor(s)
      平尾努,永田昌明,小林尚輝,奥村学
    • Industrial Property Rights Holder
      平尾努,永田昌明,小林尚輝,奥村学
    • Industrial Property Rights Type
      特許
    • Industrial Property Number
      2019-028629
    • Filing Date
      2019
    • Related Report
      2018 Annual Research Report
  • [Patent(Industrial Property Rights)] 木構造解析装置2019

    • Inventor(s)
      平尾努,永田昌明,小林尚輝,奥村学
    • Industrial Property Rights Holder
      平尾努,永田昌明,小林尚輝,奥村学
    • Industrial Property Rights Type
      特許
    • Industrial Property Number
      2019-035758
    • Filing Date
      2019
    • Related Report
      2018 Annual Research Report

URL: 

Published: 2017-04-28   Modified: 2021-02-19  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi