A RELAXATION APPROACH TO THE FUZZY CLUSTERING PROBLEM

In this paper a new algorithm for fuzzy clustering is presented. The p roposed algorithm utilizes the idea of relaxation. Convergence of the proposed algorithm is proved and limits on the relaxation parameter ar e derived. Stopping criteria and resulting convergence behaviour of th e algorithms are discussed. The performance of the new algorithm is co mpared to the fuzzy c-means algorithm by testing both on three publish ed data sets. Theoretical and empirical results reported in this paper show that the new algorithm is more efficient and leads to significan t computational savings.