It is well known that solutions of k-means algorithm depend on the initialization of cluster centers and the final solution converges to local minima. In this paper, we introduce a clustering approach that combines ideas from modified particle swarm optimization (PSO) and k-means. The potential benefits of this technique are investigated by incorporating the multidimensional asynchronism and stochastic disturbance method to the velocity in the particle swarm optimizer to create new modifications of the clustering for k-means algorithmic model, which could keep populations diversity and ability of search global optimum as well as solve the problem of the curse of dimensionality. The simulation results of web log dataset show that the proposed algorithm, compared with the five previous developed PSO techniques, provides enhanced performance and maintains more diversity in the swarm.