Multi objective line symmetry based evolutionary clustering approach

Abstract

In this chapter, we present a multi objective genetic clustering approach, in which data points are assigned to clusters based on new line symmetry based distance. Two objective functions, one based on the Euclidean distance, Davies Bouldin (DB) index, and another line symmetry distance based objective function is used. The multiple randomized kd trees based nearest neighbor search is used to reduce the complexity of finding the closest symmetric points. Experimental results based on several artificial data sets show that proposed multi objective genetic clustering can obtain optimal clustering solutions in terms of different cluster quality measures and classification performance with compared to existing clustering algorithm SBKM.