Development of statistical data analysis framework for navigation and human mobility data analysis

2020 Fiscal Year Final Research Report

• Project Area: Systems Science of Bio-navigation
• Project/Area Number: 16H06538
• Research Category: Grant-in-Aid for Scientific Research on Innovative Areas (Research in a proposed research area)
• Allocation Type: Single-year Grants
• Review Section: Complex systems
• Research Institution: Nagoya Institute of Technology
• Principal Investigator: Ichiro Takeuchi 名古屋工業大学, 工学(系)研究科(研究院), 教授 (40335146)
• Co-Investigator(Kenkyū-buntansha): 打矢 隆弘 名古屋工業大学, 工学(系)研究科(研究院), 准教授 (10375157) ; 梶岡 慎輔 名古屋工業大学, 工学(系)研究科(研究院), 助教 (40609517) ; 烏山 昌幸 名古屋工業大学, 工学(系)研究科(研究院), 准教授 (40628640)
• Project Period (FY): 2016-06-30 – 2021-03-31
• Keywords: 機械学習 / 軌跡マイニング / 選択的推論

Outline of Final Research Achievements

In this study, we established a statistical data analysis methods that can be used to analyze various moving behaviors of various animals (including humans). In this study, we introduced an approach called selective inference and to enable statistical inference of the results of animal behavior data analysis. In particular, we developed data analysis methods with statistical reliability guarantee in the tasks of extracting partial trajectories that differ among groups and extracting change points from movement trajectories. The developed data analysis methods were applied to various trajectory analysis of various animal species including humans, and its usefulness was demonstrated.

Free Research Field

機械学習

Academic Significance and Societal Importance of the Research Achievements

計測技術の発展により，車，ヒト，動物などの移動行動計測が可能となった．膨大な移動行動データを分析して知識を抽出する際には，統計的選択バイアスが生じるため，これまでは正しい信頼性評価が困難であった．本研究では，移動行動分析分野で初めて選択的推論と呼ばれる選択バイアス補正法を活用し，移動行動データから信頼性の高い知識を得る枠組を初めて開発し，その有効性を実証した．