Combining PSO and K-Means to Enhance Data Clustering


In this paper we propose a clustering method based on combination of the particle swarm optimization (PSO) and the k-mean algorithm. PSO algorithm was showed to successfully converge during the initial stages of a global search, but around global optimum, the search process will become very slow. On the contrary, k-means algorithm can achieve faster convergence to optimum solution. At the same time, the convergent accuracy for k-means can be higher than PSO. So in this paper, a hybrid algorithm combining particle swarm optimization (PSO) algorithm with k-means algorithm is proposed we refer to it as PSO-KM algorithm. The algorithm aims to group a given set of data into a user specified number of clusters. We evaluate the performance of the proposed algorithm using five datasets. The algorithm performance is compared to K-means and PSO clustering.

Meeting Name

2008 International Symposium on Telecommunications, IST 2008 (2008: Aug. 27-28, Tehran, Iran)


Electrical and Computer Engineering

Keywords and Phrases

Algorithm Performance; Articles; Clustering Methods; Data Clustering; Data Sets; Faster Convergence; Global Optimum; Global Search; Hybrid Algorithms; Initial Stages; K-Means; K-Means Algorithm; Number of Clusters; Optimum Solution; Particle Swarm Optimization Algorithm; PSO Algorithms; Search Process; Cluster Analysis; Clustering Algorithms; Convergence of Numerical Methods; Particle Swarm Optimization (PSO); Particle Swarm Optimization

International Standard Book Number (ISBN)


Document Type

Article - Conference proceedings

Document Version


File Type





© 2008 Institute of Electrical and Electronics Engineers (IEEE), All rights reserved.

Publication Date

01 Aug 2008