Data clustering based on particle swarm optimization with Lévy mechanism

Xiaoyong Liu

COMPUTER MODELLING & NEW TECHNOLOGIES 2014 18(12B) 45-49

Department of Computer Science, Guangdong Polytechnic Normal University, Guangzhou, Guangdong, 510665, China

Clustering analysis is a popular approach in data mining field. It is often used to automatically find classes or groups for unlabeled datasets. This paper looks into the use of Particle Swarm Optimization (PSO) for cluster analysis. In standard PSO, the non-oscillatory route can quickly cause a particle to stagnate and also it may prematurely converge on suboptimal solutions that are not even guaranteed to local optimal solution. In this paper, Lévy Mechanism is proposed for the particle swarm optimization (PSO) algorithm and applied in the data sets. Results show that the new PSO model, named LPSO, provides enhanced performance for clustering data.