Development of a time series analysis method for Kansei evaluation

2015 Fiscal Year Final Research Report

• Project/Area Number: 25330330
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Kansei informatics
• Research Institution: Kanazawa Institute of Technology
• Principal Investigator: Matsushita Yutaka 金沢工業大学, 情報フロンティア学部, 准教授 (60393568)
• Project Period (FY): 2013-04-01 – 2016-03-31
• Keywords: 感性情報 / 感性計測評価 / 時系列解析 / 動画像

Outline of Final Research Achievements

The aim of this research is to develop a time series analysis method for evaluation of moving images such that we can clarify subjects’ structure of consciousness. First, since interaction between shots seems to play an important role in the evaluation, we derive the conditions for using a weighted additive model, which corresponds to an auto-regression model in which the coefficients are expressed by powers of a constant. This is accomplished by generalizing extensive structures, and hence we can verify the validity of applying the model to preference comparisons between scenes consisting of any finite number of shots. Second, providing a video including several times of alternations of shots, we consider the effect of each alternation on the increment of boredom by the time series analysis. It is suggested that in the case of subjects who perceive boredom suddenly at a certain number of times of alternatives, the much previous alternations of shots give rise to this feeling.

Free Research Field

効用理論