Hierarchical Analysis Method which Deals with Data represented in Subjective/Pluralistic Form and Its Application to Group Evaluation

2015 Fiscal Year Final Research Report

• Project/Area Number: 25330323
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Kansei informatics
• Research Institution: Kyushu Institute of Technology
• Principal Investigator: HORIO Keiichi 九州工業大学, 大学院生命体工学研究科, 准教授 (70363413)
• Co-Investigator(Kenkyū-buntansha): ISOGAI Hirohisa 九州工業大学, 大学院情報工学研究院, 准教授 (70223055)
• Project Period (FY): 2013-04-01 – 2016-03-31
• Keywords: 関係データ / 定性的データ / 多元的データ / 心理尺度 / 競技成績 / データ解析

Outline of Final Research Achievements

The objective of this study is to implement an analysis method which deals with a relational data represented in subjective/pluralistic form. To achieve the objective, we investigated a data representation method and a hierarchical analysis method. As a method to analyze the group, such as team sports, psychological measure of each player and the competition results were measured, we tried to analysis based on it. Psychological measure and the competition results are quite different in expressive form. We apply a Tensor SOM which is an extension of self-organizing map (SOM) to the data acquired, and we indicate a possibility a Tensor SOM is useful for analyzing this kind of data. In addition, to analysis of effect of the psychological state on the competition results, we indicate its possibility from both of psychological and data analysis approaches.

Free Research Field

データ解析