Research on Time-series Approximation and Creation of its Applications in Sports and Health Areas

2015 Fiscal Year Final Research Report

• Project/Area Number: 24300039
• Research Category: Grant-in-Aid for Scientific Research (B)
• Allocation Type: Partial Multi-year Fund
• Section: 一般
• Research Field: Media informatics/Database
• Research Institution: Ritsumeikan University
• Principal Investigator: Kawagoe Kyoji 立命館大学, 情報理工学部, 教授 (40298724)
• Co-Investigator(Kenkyū-buntansha): Huang Hung-Hsuan 立命館大学, 情報理工学部, 准教授 (00572950)
• Project Period (FY): 2012-04-01 – 2016-03-31
• Keywords: 時系列データ / 情報検索 / 類似検索 / スポーツ応用 / エージェント / データベース

Outline of Final Research Achievements

This research aims at the improvement of the accuracy for similarity-searching of discrete multi-dimensional time series data and the development of their new applications in sports and health science areas. In this research, we proposed new methods of approximation and similarity-search for multi-dimensional time series data in an efficient and accurate way. In addition, we explored a new sport related research area by incorporating embodied virtual agent technology and constructed the guide agent who supports players by giving proper and timely advices. As the results of this research, we developed system architecture for realizing guide agents in sports with the use of multi-dimensional time series data representing the trajectories of the movements of players. We also developed its prototype systems in the sport areas of American football and ballroom dance.

Free Research Field

マルチメディア情報検索