一种改进的基于粒子群的离线聚类算法

An Improved Offline Clustering Algorithm Based on Particle Swarm

为了解决K-means算法中对于初值的敏感,提出了一种基于粒子群的改进的K-means聚类算法(IPSOFCM).在K-means算法中引入粒子群算法,可有效提高算法的全局搜索能力,有助于粒子更容易跳出局部束缚.实验结果证明,IPSOFCM算法聚类准确度高,稳定性好.

This paper puts forward a kind of improved K-means clustering algorithm based on Particle Swarm （IPSOFCM） ,in order to solve the K-means algorithm for the initial value sensitivity. Introducing the particle swarm algorithm in the K-means algorithm,which can effectively improve the algorithm＇ s global search ability and contribute to the particles are more easily jump out of local bonds. Experimental results show that the IPSOFCM clustering algorithm is with high accuracy,and good stability.