Fuzzy clustering with novel separable criterion

Abstract

Fuzzy clustering has been used widely in pattern recognition, image processing, and data analysis. An improved fuzzy clustering algorithm was developed based on the conventional fuzzy c-means (FCM) to obtain better quality clustering results. The update equations for the membership and the cluster center are derived from the alternating optimization algorithm. Two fuzzy scattering matrices in the objective function assure the compactness between data points and cluster centers, and also strengthen the separation between cluster centers in terms of a novel separable criterion. The clustering algorithm properties are shown to be an improvement over the FCM method's properties. Numerical simulations show that the clustering algorithm gives more accurate clustering results than the FCM method.