Statistical Machine Learning with Heterogeneous Auxiliary Information

• Project/Area Number: 25870322
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Statistical science; Intelligent informatics
• Research Institution: Gifu University
• Principal Investigator: SHIGA Motoki 岐阜大学, 工学部, 助教 (20437263)
• Project Period (FY): 2013-04-01 – 2016-03-31
• Project Status: Completed (Fiscal Year 2015)
• Budget Amount: ¥4,290,000 (Direct Cost: ¥3,300,000、Indirect Cost: ¥990,000); Fiscal Year 2015: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000); Fiscal Year 2014: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000); Fiscal Year 2013: ¥1,820,000 (Direct Cost: ¥1,400,000、Indirect Cost: ¥420,000)
• Keywords: 行列分解 / クラスタ解析 / スパース正則化 / 変分ベイズ学習 / 機械学習 / 非負値行列分解 / 特徴選択 / 半教師あり学習 / クラスタリング

Outline of Final Research Achievements

Recent developments of measuring engineering enable us to simultaneously monitor multiple variables and then the common size of datasets has been increasing. Thus finding essential simple rules hidden in such huge datasets becomes important. This research project has developed efficient clustering and matrix/tensor factorization methods by combining auxiliary information provided from databases and experts. Among a lot of data structures of auxiliary information, this project focused on auxiliary group and network structures. These results were also applied to data analysis on genome science, medical research, and material science.

Research Products

• [Journal Article] Two-step feature selection for predicting survival time of patients with metastatic castrate resistant prostate cancer (2016)
  • Author(s): Motoki Shiga
  • Journal Title: F1000 Research Volume: XX
  • Acknowledgement Compliant
• [Journal Article] Non-negative Matrix Factorization with Auxiliary Information on Overlapping Groups (2015)
  • Author(s): Motoki Shiga, Hiroshi Mamitsuka
  • Journal Title: IEEE Transactions on Knowledge and Data Engineering Volume: 27 Issue: 6 Pages: 1615-1628
  • DOI: 10.1109/tkde.2014.2373361
  • NAID: 120005663300
  • Peer Reviewed / Acknowledgement Compliant
• [Journal Article] Direct Conditional Probability Density Estimation with Sparse Feature Selection (2015)
  • Author(s): Motoki Shiga, Voot Tangkaratt, Masashi Sugiyama
  • Journal Title: Machine Learning Volume: 100 Pages: 161-182
  • Peer Reviewed / Acknowledgement Compliant
• [Journal Article] Variational Bayes Co-clustering with Auxiliary Information (2013)
  • Author(s): Motoki Shiga, Hiroshi Mamitsuka
  • Journal Title: Proc. of the 4th MultiClust Workshop on Multiple Clusterings, Multi-view Data, and Multi-source Knowledge-driven Clustering (MultiClust2013) Pages: 1-4
  • DOI: 10.1145/2501006.2501012
  • Peer Reviewed
• [Presentation] 走査型電子顕微鏡データ解析のためのスパース制約に基づく非負値行列分解法 (2015)
  • Author(s): 志賀元紀, 巽一厳, 津田宏治, 武藤俊介
  • Organizer: 日本金属学会 2015年秋期公園大会
  • Place of Presentation: 九州大学 伊都キャンパス, 福岡
  • Year and Date: 2015-09-16
• [Presentation] Direct Conditional Probability Density Estimation with Sparse Feature Selection (2015)
  • Author(s): Motoki Shiga, Voot Tangkaratt, Masashi Sugiyama
  • Organizer: European Conference on 25th Machine Learning and 18th Principles and Practice of Knowledge Discovery in Databases
  • Place of Presentation: Porto, Portugal
  • Year and Date: 2015-09-07
  • Int'l Joint Research
• [Presentation] 多様な特徴量に基づく目的変数予測のための統計的機械学習 (2014)
  • Author(s): 志賀元紀
  • Organizer: 日本化学会情報科学部会主催 第二回若手の会
  • Place of Presentation: 日本化学会化学会館, 東京
  • Year and Date: 2014-11-29
  • Invited
• [Presentation] 機械学習に関する最適化問題 (2014)
  • Author(s): 志賀元紀
  • Organizer: 日本オペレーションズ・リサーチ学会・中部支部研究会
  • Place of Presentation: 名古屋工業大学、名古屋
  • Year and Date: 2014-10-25
  • Invited
• [Presentation] Non-negative Matrix Factorization with Auxiliary Information (2013)
  • Author(s): Motoki Shiga, Hiroshi Mamitsuka
  • Organizer: 2013 Sapporo Workshop on Machine Learning and Applications to Biology
  • Place of Presentation: Sapporo, Japan