A New Clustering Algorithm Based on Artificial Immune System

Abstract

According the clustering principles in data analysis, a clustering algorithm based on artificial immune system is proposed in this paper. This algorithm based on the immune mechanism of the capture of antigen by the antibody. The datum that need to be clustered are viewed as antigens, and the cluster centers are viewed as the antibodies in the immune system. The clustering is effectively the process in which the immune system constantly generates antibodies for the recognition of the antigens and finally generates the optimal antibodies for the capture of the antigens. The experimental results show that the algorithm not only avoids the local optima and is robust to initialization, but also increases the convergence speed.