A machine learning based approach to analysing latent substructure of graph data

• Project/Area Number: 26730120
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Intelligent informatics
• Research Institution: Nagoya Institute of Technology (2015-2016); Kyoto University (2014)
• Principal Investigator: Karasuyama Masayuki 名古屋工業大学, 工学(系)研究科(研究院), 准教授 (40628640)
• Project Period (FY): 2014-04-01 – 2017-03-31
• Project Status: Completed (Fiscal Year 2016)
• Budget Amount: ¥3,900,000 (Direct Cost: ¥3,000,000、Indirect Cost: ¥900,000); Fiscal Year 2016: ¥780,000 (Direct Cost: ¥600,000、Indirect Cost: ¥180,000); Fiscal Year 2015: ¥1,560,000 (Direct Cost: ¥1,200,000、Indirect Cost: ¥360,000); Fiscal Year 2014: ¥1,560,000 (Direct Cost: ¥1,200,000、Indirect Cost: ¥360,000)
• Keywords: 機械学習 / グラフ / 多様体学習 / 半教師学習 / 半教師付き学習 / 多様体 / バイオインフォマティクス

Outline of Final Research Achievements

A variety of data can be represented as a graph. For statistical data analysis based on graphs, it is important to consider setting appropriate parameters of graphs and extracting informative structure. This study focuses on a methodological study of ``machine learning'' based on graph data. In particular, the label estimation problem on a graph and the important subgraph identification problem have been considered. For example, accurate label estimation methods and scalable subgraph identification methods are required by biological data analysis.

Research Products

• [Journal Article] Adaptive Edge Weighting for Graph-Based Learning Algorithms (2017)
  • Author(s): M. Karasuyama, and H. Mamitsuka,
  • Journal Title: Machine Learning Volume: 106 Issue: 2 Pages: 307-335
  • DOI: 10.1007/s10994-016-5607-3
  • Peer Reviewed / Open Access / Acknowledgement Compliant
• [Presentation] 多様体学習に基づく特徴クラスタリング (2016)
  • Author(s): 烏山昌幸
  • Organizer: 応用数理学会2016年会
  • Place of Presentation: 九州
  • Year and Date: 2016-09-12
• [Presentation] 機械学習による粒界データ解析 (2016)
  • Author(s): 烏山昌幸
  • Organizer: 応用数理学会2016年会
  • Place of Presentation: 九州
  • Year and Date: 2016-09-12
  • Invited
• [Presentation] Safe Pattern Pruning: An Efficient Approach for Predictive Pattern Mining (2016)
  • Author(s): K. Nakagawa, S. Suzumura, M. Karasuyama, K. Tsuda, and I. Takeuchi
  • Organizer: The 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
  • Place of Presentation: USA
  • Year and Date: 2016-08-13
  • Int'l Joint Research
• [Presentation] 複数多様体の同時推定に基づく特徴クラスタリング (2015)
  • Author(s): 烏山昌幸
  • Organizer: 情報論的学習理論と機械学習
  • Place of Presentation: 筑波
  • Year and Date: 2015-11-27
• [Presentation] 機械学習によるグラフデータ解析 (2015)
  • Author(s): 烏山昌幸
  • Organizer: 日本生体医工学会
  • Place of Presentation: 名古屋
  • Year and Date: 2015-05-09
  • Invited
• [Presentation] Learning Kernel-based Feature Representation for Gene Essentiality Prediction (2014)
  • Author(s): 烏山昌幸
  • Organizer: RECOMB/ISCB Conference on Regulatory and Genomics with DREAM Challenges and Cytoscape Wrokshops 2014
  • Place of Presentation: アメリカ
  • Year and Date: 2014-11-11
  • Invited
• [Presentation] Manifold-based Similarity Adaptation for Label Propagation (2014)
  • Author(s): 烏山昌幸
  • Organizer: 第17回 画像の認識・理解シンポジウム
  • Place of Presentation: 岡山
  • Year and Date: 2014-07-29
  • Invited
• [Book] 統計的学習の基礎 -データマイニング・推論・予測- (2014)
  • Author(s): 杉山 将, 井手 剛, 神嶌 敏弘, 栗田 多喜夫, 前田 英作監訳, 井尻 善久, 井手 剛, 岩田 具治, 金森 敬文, 兼村 厚範, 烏山 昌幸, 河原 吉伸, 木村 昭悟, 小西 嘉典, 酒井 智弥, 鈴木 大慈, 竹内 一郎, 玉木 徹, 出口 大輔, 冨岡 亮太, 波部 斉, 前田 新一, 持橋 大地, 山田 誠
  • Total Pages: 888
  • Publisher: 共立出版