A Maximizing Model of Bezdek-Like Spherical Fuzzyc-Means
Yuchi Kanzawa · Journal of Advanced Computational Intelligence and Intelligent Informatics · 2015
In this study, a maximizing model of Bezdek-like spherical fuzzyc-means clustering is proposed, which is based on the regularization of the maximizing model of spherical hardc-means. Such a maximizing model was unclear in Bezdek-like method, whereas other types of methods have been investigated well both in minimizing and maximizing model. Using theoretical analysis and numerical experiments, the classification rule of the proposed method is shown. Using numerical examples, the proposed method is shown to be valid for document clustering, because documents are represented as spherical data via term document-inverse document frequency weighting and normalization processing.