基于免疫粒子群优化的模糊C均值聚类算法  

Fuzzy C-means clustering algorithm based on immune particle swarm optimization

作　　者：韩琳[1] 贺兴时[1] 

机构地区：[1]西安工程大学理学院,陕西西安710048

出　　处：《西安工程科技学院学报》2007年第3期355-361,共7页Journal of Xi an University of Engineering Science and Technology

基　　金：陕西省教育厅自然科学专项基金资助项目(06JK286)

摘　　要：把免疫系统的免疫信息处理机制引入到粒子群优化(PSO)算法中,并与模糊C均值(FCM)算法相结合提出一种新的模糊聚类算法.新算法用免疫粒子群优化算法代替FCM算法的基于梯度下降的迭代过程,使算法具有较强的全局搜索能力,很大程度上避免了FCM算法易陷入局部极小的缺陷,同时也降低了FCM算法对初始值的敏感度.采用对当基思想初始化种群,获得更优的初始候选解,提高算法聚类过程中的收敛速度.以UCI机器学习数据库中的两组数据集为研究对象,实验结果表明,该算法优于基于PSO的模糊C均值聚类算法和FCM算法.By combining the properties of both Particle Swarm Optimization （PSO） algorithm in which the immune information processing mechanism of immune system is involved and Fuzzy C-Means （FCM） method, a novel fuzzy clustering algorithm is proposed. The iteration process is replaced by the PSO algorithm with immunity based on the gradient descent of FCM, which makes the algorithm have a strong global searching capacity and avoids the local minimum problems of FCM. At the same time, FCM is no longer a large degree dependent on the initialization values. Moreover, it employs opposition-based learning for population initialization to obtain fitter starting candidate solutions and improve the conver- gence speed. A real application in classifying two data sets in UCI machine learning database is provided. Numerical experiments show that the proposed algorithm is better than fuzzy c-means clustering based on PSO and FCM.

关 键 词：粒子群优化算法 模糊聚类 模糊C均值算法 免疫系统 对当基 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]