Abstract
In the field of marketing, a questionnaire is one of the most important approaches in order to design a marketing strategy. Recently, people have a variety of individualities, so respondents have various impressions to evaluation objects. In the analysis of questionnaire data, it is important not only to analyze overall trends but also to discover minority groups which have strong impressions but are different from general groups. It is, however, difficult to extract these minority groups by conventional cluster analyses. This paper tries to extract minority groups using the spectral clustering method which considers local similarity and extracts the clusters having less connection to general groups.
