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

Learning on Structure-Activity Relationship from Heterogenous Chemical Compound Databases

Research Project

Project/Area Number 17K00320
Research Category

Grant-in-Aid for Scientific Research (C)

Allocation TypeMulti-year Fund
Section一般
Research Field Intelligent informatics
Research InstitutionKwansei Gakuin University

Principal Investigator

Inokuchi Akihiro  関西学院大学, 理工学部, 教授 (70452456)

Co-Investigator(Kenkyū-buntansha) 田中 大輔  関西学院大学, 理工学部, 准教授 (60589399)
Project Period (FY) 2017-04-01 – 2021-03-31
Project Status Completed (Fiscal Year 2020)
Budget Amount *help
¥4,550,000 (Direct Cost: ¥3,500,000、Indirect Cost: ¥1,050,000)
Fiscal Year 2019: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000)
Fiscal Year 2018: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
Fiscal Year 2017: ¥1,820,000 (Direct Cost: ¥1,400,000、Indirect Cost: ¥420,000)
Keywords機械学習 / データベース / グラフ / 構造活性相関 / グラフ分類 / クラスタリング / データマイニング / 深層学習 / ケモインフォマティクス
Outline of Final Research Achievements

In this work, we investigated various approaches for learning on structure-activity relationship from heterogenous chemical compound databases. The first outcome is establishing a methodology for efficiently searching graphs contained in a query graph from a database consisting of a huge amount of graphs. Our approach is based on the prefix tree of graph codes that represent the graphs in the database. By using the prefix tree as an index, we simultaneously compute the subgraph isomorphism problem (which is known to be NP-complete) between the query graph and multiple graphs in the database. The second outcome is reducing the over smoothing phenomenon in Graph Convolution Networks in the deep learning research domain. In our approach, we combined Graph Convolution networks with the dense connection, and increased a certain percentage of prediction accuracies for various benchmark datasets compared with some conventional methods.

Academic Significance and Societal Importance of the Research Achievements

深層学習法は,現在,盛んに研究が実施されている研究分野である.その中で,様々なタイプのデータが解析されているが,グラフは非常に高い表現力を有していることが知られ,それに対するデータ解析手法や学習手法は非常に重要である.我々の成果は,データベースと機械学習の基礎研究分野への貢献であり,それらを発展させていくことで,実用化を目指せるものである.実用化の具体例の1つは,創薬化学分野である.

Report

(5 results)
  • 2020 Annual Research Report   Final Research Report ( PDF )
  • 2019 Research-status Report
  • 2018 Research-status Report
  • 2017 Research-status Report
  • Research Products

    (14 results)

All 2020 2019 2018

All Journal Article (1 results) (of which Peer Reviewed: 1 results) Presentation (10 results) (of which Int'l Joint Research: 3 results) Book (2 results) Patent(Industrial Property Rights) (1 results)

  • [Journal Article] Efficient Supergraph Search Using Graph Coding2020

    • Author(s)
      Shun IMAI, Akihiro INOKUCHI
    • Journal Title

      IEICE Transactions on Information and Systems

      Volume: E103.D Issue: 1 Pages: 130-141

    • DOI

      10.1587/transinf.2019EDP7011

    • NAID

      130007779062

    • ISSN
      0916-8532, 1745-1361
    • Year and Date
      2020-01-01
    • Related Report
      2019 Research-status Report
    • Peer Reviewed
  • [Presentation] Refining Similarity Matrices to Cluster Attributed Networks Accurately2020

    • Author(s)
      Yuta Yajima, Akihiro Inokuchi
    • Organizer
      International Workshop on GPU Computing and AI
    • Related Report
      2020 Annual Research Report
    • Int'l Joint Research
  • [Presentation] 確率的なリラベルを用いたグラフ分類の精度向上2020

    • Author(s)
      辻川拓摩,猪口明博
    • Organizer
      情報処理学会 データベースシステム研究会
    • Related Report
      2020 Annual Research Report
  • [Presentation] Generalized Multiple Kernel Learningを用いた近似グラフ編集距離の最適化による有機化合物の変異原性予測2020

    • Author(s)
      藤田将輝,猪口明博
    • Organizer
      人工知能学会 第119回 知識ベースシステム研究会
    • Related Report
      2019 Research-status Report
  • [Presentation] Attributed Network の高精度クラスタリングのための類似度行列の洗練2020

    • Author(s)
      矢嶋 悠太,猪口 明博
    • Organizer
      人工知能学会 第119回 知識ベースシステム研究会
    • Related Report
      2019 Research-status Report
  • [Presentation] アイテム密度に応じて分割されたオートエンコーダによる推薦システム2019

    • Author(s)
      谷岡豪,猪口明博
    • Organizer
      電子情報通信学会 人工知能と知識処理
    • Related Report
      2019 Research-status Report
  • [Presentation] 頂点併合と辺削除によるグラフ系列クラスタリングの効率化2019

    • Author(s)
      木村至貴,猪口明博
    • Organizer
      電子情報通信学会 人工知能と知識処理
    • Related Report
      2019 Research-status Report
  • [Presentation] 人工化合物を用いたディープラーニングによる変異原性の予測2018

    • Author(s)
      桃田侑典,猪口明博
    • Organizer
      人工知能学会第114回 知識ベースシステム研究会
    • Related Report
      2018 Research-status Report
  • [Presentation] 交差検定を用いた説明属性の選択によるLocal SVMの正答率向上2018

    • Author(s)
      松井勇大,猪口明博
    • Organizer
      人工知能学会第114回 知識ベースシステム研究会
    • Related Report
      2018 Research-status Report
  • [Presentation] Accurate and Fast Computation of Approximate Graph Edit Distance based on Graph Relabeling2018

    • Author(s)
      Sousuke Takami and Akihiro Inokuchi
    • Organizer
      the 7th International Conference on Pattern Recognition Applications and Methods
    • Related Report
      2017 Research-status Report
    • Int'l Joint Research
  • [Presentation] Transition-based dependency parser with postponed determinations for Japanese sentences.2018

    • Author(s)
      Xiaobo Xi and Akihiro Inokuchi
    • Organizer
      International Conference on Asian Language Processing
    • Related Report
      2017 Research-status Report
    • Int'l Joint Research
  • [Book] Pattern Recognition Applications and Methods: 6th International Conference, ICPRAM 2017. Chapter 2, Graph Classification with Mapping Distance Graph Kernels2018

    • Author(s)
      Maria De Marsico, Gabriella Sanniti di Baja, Ana Fred (編), Tetsuya Kataoka(著), Eimi Shiotsuki(著), Akihiro Inokuchi(著)
    • Total Pages
      252
    • Publisher
      Springer
    • ISBN
      9783319936468
    • Related Report
      2018 Research-status Report
  • [Book] Learning from Data Streams in Evolving Environments: Methods and Applications. Chapter 10, Detecting Smooth Cluster Changes in Evolving Graph2018

    • Author(s)
      Moamar Sayed-Mouchaweh (編), Sohei Okui, Kaho Osamura, and Akihiro Inokuchi
    • Total Pages
      317
    • Publisher
      Springer
    • ISBN
      3319898027
    • Related Report
      2018 Research-status Report
  • [Patent(Industrial Property Rights)] 人工化合物データを用いた化合物特性予測の深層学習法 および装置,並びに,化合物特性予測方法および装置2018

    • Inventor(s)
      桃田侑典,猪口明博
    • Industrial Property Rights Holder
      桃田侑典,猪口明博
    • Industrial Property Rights Type
      特許
    • Filing Date
      2018
    • Related Report
      2018 Research-status Report

URL: 

Published: 2017-04-28   Modified: 2022-01-27  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi