New Information-based Clustering Method using Renyi's Entropy and Fuzzy C-Means Clustering

M. Aghagolzadeh (Iran), H. Soltanian-Zadeh (Iran/USA), and B.N. Araabi (Iran)

Keywords

Information theory, Renyi’s entropy, Topdown hierarchical algorithms, clustering.

Abstract

: This paper presents a new clustering method based on Renyi entropy. The proposed method maximizes entropy of clusters using between and within clusters entropies. It is a top-down multi-resolution method and uses the initial clusters found by Fuzzy C Means. Applications of the proposed algorithm on the synthetic data are compared with those of C-Means and Gustafson-Kessel algorithms. Results show superiority of the proposed algorithm to these methods.

Important Links:



Go Back