An Efficient Algorithm for Solving Data Clustering Problems

This paper presents a new data clustering algorithm called KPSO algorithm, a combination on K-means and Particle swarm Optimization algorithms. Unlike traditional K-means method, KPSO need not specify the number of clusters to be given prior the clustering process and is able to find the optimal number of clusters during the clustering process. In each and every iteration of KPSO, a discrete PSO is used to optimize the number of clusters with which the K-means is used to find the best clustering result.