针对K均值聚类算法在全局优化中的不足,提出了基于粒子群的K均值(PSO-KM)聚类算法。粒子群优化算法作为一种基于群智能方法的演化计算技术,有很好的全局搜索能力。通过理论分析及实验证明,该算法有较好的全局收敛性,能有效地克服传统的K均值算法易陷入局部极小值的缺点。对KDD-99数据集的仿真实验结果表明,该算法在入侵检测中能获得令人满意的检测率和误检率。 Aiming at the shortage of K-means clustering algorithm in global optimization, a K-means clustering algorithm based on particle swarm optimization (PSO-KM) is proposed. Particle Swarm Optimization (PSO), as an evolutionary computation technique based on swarm intelligence, has good global search ability. The theoretical analysis and experimental results show that the proposed algorithm has better global convergence and can effectively overcome the shortcomings of the traditional K-means algorithm into local minima. Simulation results on KDD-99 dataset show that this algorithm can achieve satisfactory detection rate and false positive rate in intrusion detection.