For the shortcomings of K-means clustering algorithm,a new clustering algorithm is introduced, and it integrates particle swarm optimization(PSO) and K-means algorithm. In the new algorithm,it first u-ses particle swarm optimization and K-means algorithm to search global optimum location, and then K-means algorithm is used for rapidly finding optimal cluster centers in the global optimal solution space. After testing four data sets of experiment, the algorithm is compared with the K-means algorithm, particle swarm-based K-means algorithm. The experimental results show that the quality of the new clustering algorithm is better than latter two algorithms.%针对K-均值聚类算法存在的不足,提出了一种新的整合粒子群优化算法(PSO)和K-均值算法的聚类算法.在新算法中,首先结合使用粒子群优化算法和K-均值算法搜索全局最优解的位置,然后再用K-均值算法在全局最优解附近的局部空间内快速寻找最优聚类中心.通过对4个数据集的实验测试,将此算法与K-均值算法、基于粒子群的K-均值算法进行了比较.实验结果表明,新算法的聚类质量比后两个算法更优.
展开▼
