首页 | 本学科首页   官方微博 | 高级检索  
     

人工免疫C-均值聚类算法
引用本文:张雷,李人厚. 人工免疫C-均值聚类算法[J]. 西安交通大学学报, 2005, 39(8): 836-839
作者姓名:张雷  李人厚
作者单位:1. 西安交通大学系统工程研究所,710049,西安;河南科技大学电子与信息工程学院,471003,洛阳
2. 西安交通大学系统工程研究所,710049,西安
摘    要:通过借鉴生物免疫系统中的克隆选择原理和记忆机制,提出了一种人工免疫C-均值混合聚类算法.该算法采用了新的克隆选择方法,通过亲和度排序和个体浓度定义了个体的选择概率,从而可确定个体的适应值评价函数,以评价和选择个体.算法还集成了一种C-均值搜索算子,用于加快收敛速度.在聚类数目已知的情况下,所提算法能够得到给定数据集下的全局最优划分,与基于遗传算法的聚类方法比较,它具有更快的收敛速度和更高的收敛精度,并可扩展到性能指标能够表示为优化聚类中心函数的聚类模型之中.仿真结果表明,所提算法是有效性的.

关 键 词:聚类算法 人工免疫 C-均值
文章编号:0253-987X(2005)08-0836-04
收稿时间:2004-10-25
修稿时间:2004-10-25

Artificial Immune C-Means Clustering Algorithm
Zhang Lei,Li Renhou. Artificial Immune C-Means Clustering Algorithm[J]. Journal of Xi'an Jiaotong University, 2005, 39(8): 836-839
Authors:Zhang Lei  Li Renhou
Abstract:Inspired by the clone selection principle and memory mechanism of the vertebrate immune system, a hybrid algorithm combining C-means algorithm and artificial immune algorithm is presented. A new clone selection strategy is used and the individual selection probability is defined through sorting the affinity and individuals concentration so that the evaluating function of the individual fitness can be determined,and then individuals are evaluated and selected. The C-means algorithm is treated as a new search operator in order to improve the convergence speed. Comparing with the genetic algorithm based clustering approaches the proposed algorithm can converge to the global optimum faster and has higher accuracy. Given the cluster number, the algorithm can obtain the best partition of data sets. The algorithm can be extended to other clustering model whose objective function can be represented in terms of optimization of cluster centers. Experimental results indicate the validity of the proposed algorithm.
Keywords:clustering algorithm    artificial immune    C-means
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号