2009 Volume 8 Issue 4 Pages 1185-1191
Recently, there are several kinds of values with the diversification of individuality in market. Some of the values supported by few people, which is called “minority group”, at present have possibility to grow to majority groups in future with the change of historical background or people's sensitivity. It is effective and important for the market analysis to find out these minority groups at early stage. Companies often carry out questionnaires to develop marketing strategy or design new products, which is a chance to find them. In conventional methods, respondents of a questionnaire are classified based on the attributes such as gender, age and so on, and then the classified groups are analyzed or compared one another. Though the conventional analysis is effective to grasp the overall tendency of the evaluation data, it is difficult to find out minority groups because of the diversity of individuality. On the other hand, the authors have proposed the clustering methods based on the tendency of the answer of the questionnaire. This paper proposes a new method to visualize the evaluated data based on both the evaluation value and the correlation of them and to cluster respondents interactively on the visible space. This paper applies the proposed method to a web questionnaire data and shows that this method effectively supports to find out minority groups with the analysis for them.