X-SPA: Spatial Characteristic PSO Clustering Algorithm with Efficient Estimation of the Number of Cluster

Clustering is one of main technical of data mining, by a kind of non-teacher supervises recognition pattern. Despite its popularity for general clustering, K-means suffers two major shortcomings: the number of clusters K has to be supplied by the user and the search is prone to local minima. This article unifies particle swarm optimization (PSO) algorithm and Bayesian information criterion (BIC), proposes a numeric clustering algorithm. Chaos and space characteristic ideas are involved in the algorithm to avoid local optimal problem. Furthermore, BIC is also involved to provide an efficient estimation of the number of cluster. The simulation experiments indicated that, the articlepsilas algorithm has good performance in the numeric attribute cluster problems both in clustering result and estimation of K.