基于PSO的模糊K-Prototypes聚类

Fuzzy K-Prototypes clustering based on particle swarm optimization

模糊K-Prototypes(FKP)算法能够对包含数值属性和分类属性相混合的数据集进行有效聚类,但是存在对初始值敏感、容易陷入局部极小值的问题。为了克服该缺点,提出了一种基于粒子群优化(PSO)算法和FKP算法的混合聚类算法,先利用PSO算法确定FKP的初始聚类中心,再将PSO聚类结果作为后续FKP算法的初始值。实验结果表明,新算法具有良好的收敛性和稳定性,聚类效果优于单一使用FKP算法。

Fuzzy K-Prototypes （FKP） algorithm is efficient in clustering data sets with mixed numeric and categorical values, with the defects including sensitivity to the initial data and being easy to run into the local optimization. In order to overcome them, a new hybrid clustering algorithm based on particle swarm （PSO） optimization and FKP algorithm is proposed, by using PSO to determine the centroids of clusters and taking the clustering result of PSO as the initialized value of the FKP. The results show that the proposed algorithm is superior to FKP algorithm with a better astringency and stablility.