Clustering is a useful tool to explore data structures and have been employed in many disciplines. One of the most used techniques for clustering is based on K-means such that the data is partitioned into K clusters. Although K-means algorithm is easy to implement and works fast in most situations, it suffers from several drawbacks due to its choice of initializations and convergence to local optima. The K-harmonic means clustering solves the problem of initialization, but for the convergence to local optima, the K-harmonic means is hopeless. In this paper, a new method is proposed to solve the problem of convergence to local optima, namely particle swarm optimization K-harmonic means clustering (PSOKHM) algorithm. The experiment results on the three well known datasets show the effectiveness of the PSOKHM clustering algorithm.
K-Harmonic Means Data Clustering with PSO Algorithm
Adv.Intel.,Soft Computing
01.01.2012
7 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Improved K-Means Clustering Algorithm
IEEE | 2008
|Modified K-Means Clustering Algorithm
IEEE | 2008
|Artificial Immune C-Means Clustering Algorithm
British Library Online Contents | 2005
|A Robust Fuzzy c-Means Clustering Algorithm for Incomplete Data
Springer Verlag | 2017
|Modified Global K'-means Algorithm and Its Application to Data Clustering
British Library Online Contents | 2011
|