Integrated analysis for heterogeneous data from multiple sources

• Project/Area Number: 15K12112
• Research Category: Grant-in-Aid for Challenging Exploratory Research
• Allocation Type: Multi-year Fund
• Research Field: Soft computing
• Research Institution: Advanced Telecommunications Research Institute International
• Principal Investigator: KAWANABE Motoaki 株式会社国際電気通信基礎技術研究所, 脳情報通信総合研究所, 主幹研究員 (30272389)
• Co-Investigator(Kenkyū-buntansha): 兼村 厚範 国立研究開発法人産業技術総合研究所, 人間情報研究部門, 研究員 (50580297)
• Co-Investigator(Renkei-kenkyūsha): KANAI Ryota 株式会社アラヤ・ブレイン・イメージング, 代表取締役兼Chief Scientist (80607177)
• Project Period (FY): 2015-04-01 – 2017-03-31
• Project Status: Completed (Fiscal Year 2016)
• Budget Amount: ¥3,510,000 (Direct Cost: ¥2,700,000、Indirect Cost: ¥810,000); Fiscal Year 2016: ¥1,690,000 (Direct Cost: ¥1,300,000、Indirect Cost: ¥390,000); Fiscal Year 2015: ¥1,820,000 (Direct Cost: ¥1,400,000、Indirect Cost: ¥420,000)
• Keywords: 確率的情報処理 / ライフログ / 異種データの統合解析 / 確率的情報書類

Outline of Final Research Achievements

We developed machine learning techniques to infer continuously personality traits and psychological measures useful for mental/physical health based on heterogeneous data from multiple sources in real environments. As factors affecting on heath status, we focused on attention levels and behavioural patterns. For the former, we measured simultaneously EEG and biomedical signals from several subjects during an auditory sustained attention response task (SART), and extracted features related to attention levels such as ECG R-R intervals. For the latter, we developed a tensor decomposition method to infer behavioural patterns efficiently from mobility traces. Furthermore, in order to make full use of heterogeneous multi-sensor data with complex missing patterns, we proposed completion methods of missing observations based on matrix factorisation or AR models, and improvements of propensity score estimation with deep neural networks.

Research Products

• [Journal Article] Group sparsity tensor factorization for re-identification of open mobility trace (2017)
  • Author(s): Murakami, T., Kanemura, A., and Hino, H.
  • Journal Title: IEEE Transactions on Information Forensics and Security Volume: 12 (3) Issue: 3 Pages: 689-704
  • DOI: 10.1109/tifs.2016.2631952
  • Peer Reviewed / Open Access / Acknowledgement Compliant
• [Journal Article] Key frame extraction from first-person video with multi-sensor integration (2017)
  • Author(s): Li, Y., Kanemura, A., Asoh, H., Miyanishi, T., and Kawanabe, M.
  • Journal Title: Proceedings of the IEEE International Conference on Multimedia and Expo (ICME) Volume: -
  • Peer Reviewed / Acknowledgement Compliant
• [Journal Article] Modeling the propensity score with statistical learning (2016)
  • Author(s): K. Uchihashi and A. Kanemura
  • Journal Title: Lecture Notes in Computer Science Volume: 9950 Pages: 261-269
  • DOI: 10.1007/978-3-319-46681-1_32
  • ISBN: 9783319466804, 9783319466811
  • Peer Reviewed / Acknowledgement Compliant
• [Journal Article] 能力テストにおける非負行列分解を用いた適応型出題の精度解析 (2016)
  • Author(s): 兼村厚範，畑中哲生，鹿内学
  • Journal Title: 信学技報NC Volume: 115
  • Acknowledgement Compliant
• [Journal Article] Automatic categorization of health indices for risk quantification (2015)
  • Author(s): Kanemura, A., Lipowski, G., Komine, H., and Akaho, S.
  • Journal Title: Procedia Computer Science Volume: 63 Pages: 325-331
  • DOI: 10.1016/j.procs.2015.08.350
  • Peer Reviewed / Int'l Joint Research / Acknowledgement Compliant
• [Journal Article] Group sparsity tensor factorization for de-anonymization of mobility traces (2015)
  • Author(s): Murakami, T., Kanemura, A., and Hino, H.
  • Journal Title: The 14th IEEE Int. Conf. on Trust, Security and Privacy in Computing and Communications (IEEE TrustCom-15) Volume: 1 Pages: 621-629
  • DOI: 10.1109/trustcom.2015.427
  • Peer Reviewed / Acknowledgement Compliant
• [Journal Article] 能力テスト得点の非負行列分解 (2015)
  • Author(s): 兼村厚範，大成弘子，鹿内学，橋本将崇，赤穂昭太郎
  • Journal Title: 信学会技報IBISML Volume: 115
  • Acknowledgement Compliant
• [Presentation] Key frame extraction from first-person video with multi-sensor integration (2017)
  • Author(s): Li, Y., Kanemura, A., Asoh, H., Miyanishi, T., and Kawanabe, M.
  • Organizer: IEEE International Conference on Multimedia and Expo (ICME)
  • Place of Presentation: Hong Kong, China
  • Year and Date: 2017-07-10
  • Int'l Joint Research
• [Presentation] 繰り返し推定による脳波時系列の欠測補完 (2016)
  • Author(s): 鄭郁森，兼村厚範，福永修一
  • Organizer: 情報論的学習理論ワークショップ（IBIS）
  • Place of Presentation: 京都
  • Year and Date: 2016-11-17
• [Presentation] Modeling the propensity score with statistical learning (2016)
  • Author(s): Uchihashi, K., and Kanemura, A.
  • Organizer: International Conference on Neural Information Processing (ICONIP)
  • Place of Presentation: Kyoto, Japan
  • Year and Date: 2016-10-18
  • Int'l Joint Research
• [Presentation] ブレイン・マシン・インタフェース (2016)
  • Author(s): 川鍋一晃
  • Organizer: 公益社団法人日本磁気学会 第4回岩崎コンファレンス 「医工学とビッグデータが拓く医療の未来」
  • Place of Presentation: 東京
  • Year and Date: 2016-05-17
  • Invited
• [Presentation] 能力テストにおける非負行列分解を用いた適応型出題の精度解析 (2016)
  • Author(s): 兼村厚範，畑中哲生，鹿内学
  • Organizer: 通信学会NC研究会
  • Place of Presentation: 東京
  • Year and Date: 2016-03-23
• [Presentation] 能力テスト得点の非負行列分解 (2015)
  • Author(s): 兼村厚範，大成弘子，鹿内学，橋本将崇，赤穂昭太郎
  • Organizer: 情報論的学習理論ワークショップ（IBIS）
  • Place of Presentation: つくば
  • Year and Date: 2015-11-27
• [Presentation] 多変量自己回帰モデルを利用した脳波時系列の欠測値推定 (2015)
  • Author(s): 金子拓光，兼村厚範，福永修一
  • Organizer: 情報論的学習理論ワークショップ（IBIS）
  • Place of Presentation: つくば
  • Year and Date: 2015-11-18
• [Presentation] On possibility and difficulties in adaptive brain-computer interface in real environments. Germany-Japan Workshop on Adaptive BCI, Kyoto Univ., Japan, 2015. (2015)
  • Author(s): Kawanabe, M.
  • Organizer: Germany-Japan Workshop on Adaptive BCI
  • Place of Presentation: Kyoto, Japan
  • Year and Date: 2015-10-28
  • Int'l Joint Research / Invited
• [Presentation] Automatic categorization of health indices for risk quantification (2015)
  • Author(s): Kanemura, A., Lipowski, G., Komine, H., and Akaho, S.
  • Organizer: International Conference on Current and Future Trends of Information and Communication Technologies in Healthcare (ICTH)
  • Place of Presentation: Berlin, Germany
  • Year and Date: 2015-09-27
  • Int'l Joint Research
• [Presentation] Group sparsity tensor factorization for de-anonymization of mobility traces (2015)
  • Author(s): Murakami, T., Kanemura, A., and Hino, H.
  • Organizer: The 14th IEEE Int. Conf. on Trust, Security and Privacy in Computing and Communications (IEEE TrustCom-15)
  • Place of Presentation: Helsinki, Finland
  • Year and Date: 2015-08-20
  • Int'l Joint Research