The main subject of this book is the fuzzy c-means proposed by Dunn and Bezdek and their variations including recent studies. A main reason why we concentrate on fuzzy c-means is that most methodology and application studies in fuzzy clustering use fuzzy c-means, and hence fuzzy c-means should be considered to be a major technique of clustering in general, regardless whether one is interested in fuzzy methods or not. Unlike most studies in fuzzy c-means, what we emphasize in this book is a family of algorithms using entropy or entropy-regularized methods which are less known, but we consider the entropy-based method to be another useful method of fuzzy c-means. Throughout this book one of our intentions is to uncover theoretical and methodological differences between the Dunn and Bezdek traditional method and the entropy-based method. We do note claim that the entropy-based method is better than the traditional method, but we believe that the methods of fuzzy c-means become complete by adding the entropy-based method to the method by Dunn and Bezdek, since we can observe natures of the both methods more deeply by contrasting these two.
