A validity-guided support vector clustering algorithm for identification of optimal cluster configuration

Abstract

This paper presents a validity-guided support vector clustering (SVC) algorithm for identifying an optimal cluster configuration. Since the SVC is a kernel based clustering approach, the parameter of kernel functions plays a crucial role in the clustering result. Without a priori knowledge of data sets, a validity measure, based on a ratio of overall cluster compactness to separation, has been developed to automatically determine a suitable parameter of the kernel functions. Using this parameter, the SVC algorithm is capable of identifying the optimal cluster number with compact and smooth arbitrary-shaped cluster boundaries. Computer simulations have been conducted to demonstrate the effectiveness of the proposed validity-guided SVC algorithm.