一种基于免疫遗传算法的聚类方法  

A Clustering Method Based on Immune Genetic Algorithm

作　　者：曹永春[1] 邵亚斌[1] 田双亮[1] 蔡正琦[1] 

机构地区：[1]西北民族大学数学与计算机科学学院,甘肃兰州730030

出　　处：《广西师范大学学报（自然科学版）》2013年第3期59-64,共6页Journal of Guangxi Normal University:Natural Science Edition

基　　金：国家自然科学基金资助项目(11161041);2012年度国家民委科研基金资助项目;中央高校基本科研基金资助项目(31920130009)

摘　　要：为了克服遗传算法未成熟收敛现象,提高聚类效果和性能,本文提出一种基于免疫遗传算法的聚类方法。该方法将免疫原理引入遗传聚类算法中,通过对选择算子在依据适应度选择的机制上,增加基于浓度的调节因子来调整抗体的选择概率,从而保持了群体多样性,有效克服了未成熟收敛现象。算法使用改进的分组编码方式并针对该编码方式制定了相应的种群初始化方案,设计了合理的遗传算子,使算法具有自动获取最优聚类数和最优聚类方案的能力。理论分析和实验结果证明,该算法具有良好的性能和聚类效果。In order to overcome premature convergence of genetic algorithm and derive better performance and higher accuracy on clustering problems,a new clustering method based on immune genetic algorithm is proposed in this paper. Immune principle is introduced into genetic clustering algorithm in the method, through adding density regulatory factor on selection operator performed according to the fitness in the mechanism. Selection probability is adjusted, and the diversity of individuals is maintained ,and then the disadvantage of premature convergence is overcome. The algorithm represents individuals by improved grouping coding ; based on it,a corresponding manner of initial population is formulated and appropriate genetic operators are designed, which enable the algorithm automatically finding the proper number of clusters and the proper partition from a given data set. show that the algorithm derives better performance and The theoretical analysis and experiment result higher accuracy on clustering problems.

关 键 词：免疫原理 遗传算法 聚类方法 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]