Authors: Yugal Kumar; Gadadhar Sahoo
Addresses: Department of Computer Science and Engineering, Birla Institute of Technology, Mesra, Ranchi, Jharkhand, India ' Department of Computer Science and Engineering, Birla Institute of Technology, Mesra, Ranchi, Jharkhand, India
Abstract: Clustering is a popular data analysis technique to find the hidden relationship between patterns. It is an unsupervised technique which is applied to obtain the optimal cluster centres especially in partitioned-based clustering algorithms. Cat swarm optimisation (CSO) is a new meta-heuristic algorithm which has been applied to solve various optimisation problems and it provides better results in comparison to same class of algorithms. But, this algorithm suffers from diversity and local optima problems. This study presents a new hybrid CSO algorithm by incorporating Monte Carlo-based search equation and population centroid (PopC) operator with Gaussian probability distribution. Monte Carlo-based search equation is applied to enhance the diversity of the CSO algorithm. The population centroid (PopC) operator is used to prevent the CSO algorithm from trapping in local optima. The performance of the proposed algorithm is tested with several artificial and real datasets and compared with existing methods like K-means and PSO. The experimental results indicate that the proposed algorithm is a robust and effective algorithm to handle the clustering problems.
Keywords: cat swarm optimisation; CSO; Gaussian probability distribution; data clustering; Monte Carlo search equation; particle swarm optimisation; PSO; metaheuristics; swarm intelligence; population centroid operator.
International Journal of Computational Science and Engineering, 2017 Vol.14 No.2, pp.198 - 210
Available online: 13 Mar 2017 *Full-text access for editors Access for subscribers Purchase this article Comment on this article