基于QPSO的模糊C均值聚类算法

针对模糊C均值(FCM)聚类算法存在的缺点,利用量子粒子群优化(QPSO)算法的全局搜索能力,提出了一种新的聚类算法——基于量子粒子群优化的FCM聚类算法(QPSOFCM).QPSOFCM算法先对随机初始点利用QPSO进行优化,然后利用产生的中心点进行聚类,重复上述两步操作直至结果满意为止.新算法可以降低FCM算法对初始点的敏感度,一定程度上避免了FCM算法易陷入局部极优的缺陷.几组数据实验结果表明,与FCM和PSOFCM算法相比,提出的QPSOFCM算法聚类结果更可靠.

Fuzzy C-Means Clustering Based on Quantum-Behave Particle Swarm Optimization Algorithm

After analyzing the disadvantages of the Fuzzy C-means(FCM) clustering algorithm,this paper proposes a novel Fuzzy C-means clustering based on Quantum-behave Particle Swarm Optimization algorithm(QPSOFCM).Firstly we use QPSO to optimize initial centeral points,and then use FCM to cluster,and lastly iterate above two steps until gaining satisfying results.Our QPSOFCM algorithm may lower the influence of FCM clustering results by initialized data points,and avoid the advantage of easily entering local minimum.Several data experimental results show that the proposed algorithm is more robust and accurate than FCM and PSOFCM algorithm.