遗传K均值方法在品种资源分类中的应用  

The Application on Variety Clustering of Genetic-Algorithm-K-Means

作　　者：徐永春[1] 张森文[2] 

机构地区：[1]华南农业大学工程学院,广东广州510642 [2]暨南大学应用力学研究所,广东广州510632

出　　处：《华南农业大学学报》2009年第2期97-100,共4页Journal of South China Agricultural University

摘　　要：采用实数编码方式,对聚类的中心矩阵进行编码,通过数组变换将染色体与相应聚类中心的数组进行匹配,通过轮赌选择和自适应的交叉、变异操作及均值小生境的种群优化对聚类中心的编码进行更新迭代,最终得到稳态的聚类误差函数和划分效果最好的聚类中心.然后通过对某基地的甘蔗品种进行分析、比较,分析的误差函数结果显示,RingaK-Means改进的聚类效果明显优于传统的K-Means方法及Sga-K-Means方法的聚类效果.This paper proposed one kind of K-Means analysis method based on the genetic algorithm by the average value niche. The real number method was used to encode the clustering center, and the chromosome was matched with the clustering central array correspondingly throuth array transformation. The method was used to update the clustering central with selection, crossover, mutation and the average value niche population optimization. The stable state of the error function and the best dividing clustering center was obtained. The experimental result demonstrated that the Ringa-K-Means method was obviously better than the traditional K-Means method and the Sga-K-Means method.

关 键 词：遗传算法 K均值聚类 实数编码 甘蔗品种 小生境 

分 类 号：TP306.1[自动化与计算机技术—计算机系统结构]