The improved K-means clustering algorithm using the proposed extended PSO algorithm

Nowadays, clustering has a major role in the most of research areas such as engineering, medical sciences, biology and data analysis. A popular clustering algorithm is K-means. This algorithm has advantages such as high speed and ease of employment, but it suffers from the problem of local optimal. In order to overcome this problem, a lot of Research has been done in clustering. In this paper we proposed an Extended Chaotic Particle Swarm Optimization algorithm which is called ECPSO, for optimum clustering. The proposed algorithm is an extended version from standard PSO, In ECPSO Algorithm; we have enhanced the operators in the classical version of the PSO algorithm. As an example, for increased dispersion of first population, the production of the initial population is based on a Chaos Trail whereas in the classical version, it is based on randomized trial. Moreover we used from Chaos Trail in migration operator and also used from several global best for particle migration instead use a globalbest in standard PSO. The suggested method is evaluated on several standard data sets at UCI database and its performance is compared with those of Particle Swarm Optimization (PSO), Chaotic Particle Swarm Optimization (CPSO), Genetic Algorithm (GA), and hybrid algorithm PSO+K-means. The results obtained are compared with some other approaches in terms of Purity degree, Total distance within the cluster and between cluster (Intra/Inter), Convergence rate and Time complexity. The simulation results show that the proposed algorithm is capable of yielding the optimized solution with higher purity degree and better quality results in comparison to the other compared algorithms.