Fuzzy Multiset Model for Information Retrieval and Clustering Using a Kernel Function

Abstract

A method of fuzzy clustering based on a fuzzy multiset model is proposed. Data clustering has frequently been discussed in relation to information retrieval models and on the other hand fuzzy multisets provide an appropriate model of information retrieval on the WWW. A dissimilarity measure on fuzzy multisets is proposed. A method of fuzzy c-means using this measure is studied and fuzzy c-means clustering using a kernel function employed in nonlinear transformation into a high dimensional feature space in Support Vector Machine is discussed.