Development and Application of Statistical Estimation and Simulation for Super High Dimensional Data Space

2016 Fiscal Year Final Research Report

• Project/Area Number: 25240036
• Research Category: Grant-in-Aid for Scientific Research (A)
• Allocation Type: Single-year Grants
• Section: 一般
• Research Field: Intelligent informatics
• Research Institution: Osaka University
• Principal Investigator: Washio Takashi 大阪大学, 産業科学研究所, 教授 (00192815)
• Co-Investigator(Kenkyū-buntansha): 伊庭 幸人 統計数理研究所, 大学共同利用機関等の部局等, 教授 (30213200) ; Michael E.Houle 国立情報学研究所, 大学共同利用機関等の部局等, 教授 (90399270)
• Co-Investigator(Renkei-kenkyūsha): SHIMIZU Shohei 滋賀大学, データサイエンス学部, 准教授 (10509871) ; KAWAHARA Yoshinobu 大阪大学, 産業科学研究所, 准教授 (00514796) ; INOGUCHI Akihiro 関西学院大学, 理工学部, 准教授 (70452456)
• Research Collaborator: Ting Kai Ming Federation University Australia, Faculty of Science and Technology, Professor
• Project Period (FY): 2013-04-01 – 2017-03-31
• Keywords: 超高次元データ / 機械学習 / データマイニング / 人工知能 / 次元の呪い / シミュレーション / 希少事象

Outline of Final Research Achievements

In this study, we aimed to develop (1) generic and robust principles of statistical estimation and scenario generation against super high dimensionality, (2) statistical estimation methods using super high dimensional data, (3) probabilistic scenario generation methods for super high dimensional space, (4) an application of these developed methods and simulation techniques, and (5) an international research community.
Throughout this project, we developed techniques of similarity measure, density evaluation, robust estimation, scenario search, retrieval and clustering, classification, anomaly detection, rare scenario generation, and frequent pattern derivation. We also organized two international conferences and seven international workshops/seminars.

Free Research Field

機械学習