An application of particle swarm optimization algorithm to clustering analysis

Authors:
R. J. Kuo;M. J. Wang;T. W. Huang
Affiliations:
National Taiwan University of Science and Technology, Department of Industrial Management, 106, Taipei, Taiwan, ROC;National Taipei University of Technology, Department of Industrial Engineering and Management, 106, Taipei, Taiwan, ROC;National Taipei University of Technology, Department of Industrial Engineering and Management, 106, Taipei, Taiwan, ROC
Venue:
Soft Computing - A Fusion of Foundations, Methodologies and Applications - Special issue on Digital Information Forensics
Year:
2011

Citing 0
Cited 5

Application of a hybrid of genetic algorithm and particle swarm optimization algorithm for order clustering

Decision Support Systems
A survey: hybrid evolutionary algorithms for cluster analysis

Artificial Intelligence Review
Application of particle swarm optimization and perceptual map to tourist market segmentation

Expert Systems with Applications: An International Journal
Towards recommender system using particle swarm optimization based web usage clustering

PAKDD'11 Proceedings of the 15th international conference on New Frontiers in Applied Data Mining
A powerful hybrid clustering method based on modified stem cells and Fuzzy C-means algorithms

Engineering Applications of Artificial Intelligence

Quantified Score

Hi-index	0.00

Visualization

Abstract

Particle swarm optimization algorithm (PSOA), which maintains a population of particles, where each particle represents a potential solution to an optimization problem, is a population-based stochastic search process. This study intends to integrate PSOA with K-means to cluster data. It is shown that PSOA can be employed to find the centroids of a user-specified number of clusters. The proposed PSOA is evaluated using four data sets, and compared to the performance of some other PSOA-based methods and K-means method. Computational results show that the proposed method has much potential. A real-world problem for order clustering also illustrates that the proposed method is quite promising.