K-means algorithm based on particle swarm optimization for web document clustering

K-means as a clustering algorithm has been studied in Web document clustering. However, with the deficiency of global search ability it is not satisfactory. Particle swarm optimization (PSO) is one of the evolutionary computation techniques based on swarm intelligence, which has high global search ability. So K-means algorithm based on PSO (PSO-KM) was proposed in this paper. The vector space model (VSM) was employed to represent the documents, and Compressed Sparse Row (CSR) format was implemented to store the data. Computational experiments were conducted to test the performance of the hybrid algorithm using three web document datasets. The F-measure and the entropy were adopted to evaluate the quality of clustering. The results were compared with that of K-means algorithm and that of K-means algorithm based on genetic algorithm (GA-KM), which show that the quality of the clustering solutions obtained from PSO-KM is better than that from K-means or GA-KM. The run time of PSO-KM is less than that of GA-KM algorithm.