基于云自适应遗传算法的K-means聚类分析  

K-means Clustering Analysis Based on Cloud Adaptive Genetic Algorithm

作　　者：许茂增[1] 余国印 

机构地区：[1]重庆交通大学管理学院,重庆400074

出　　处：《数学的实践与认识》2015年第17期48-55,共8页Mathematics in Practice and Theory

基　　金：教育部人文社科研究项目西部地区物流业带动现代服务业的效应及动力机制(10XJA790009)

摘　　要：现有的基于遗传算法的K-means聚类算法,利用遗传算法的全局优化性提高了K-means算法的寻优能力,收敛速度却过慢.为了解决上述问题,提出基于云自适应遗传算法的K-means聚类算法,利用云模型云滴的随机性和稳定趋向性设计遗传算法的交叉和变异概率,并在进化过程中引入K均值算子,以克服算法收敛速度过慢的问题.实验比较表明,算法具有较好的全局优化性,且收敛速度较快,提高了聚类算法解决物流管理中数据聚类工作的能力.The existing K-means clustering algorithms based on genetic algorithm effectively improve the optimization capacity of K-means by using the global optimization function of the genetic algorithm, however, the convergence speed of them is quite slow. To solve this problem, we propose a K-means clustering algorithm based on the cloud adaptive genetic algorithm, which designs the probability of crossover and mutation of the genetic algorithm by using the randomness and stable tendency of cloud droplet in the cloud model, at the same time, K- means operator is introduced in evolution process. The comparison results of the experiments show that this algorithm, with better global optimization capacity and faster convergence speed, improves the working ability of data clustering by using clustering algorithm in solving logistic management.

关 键 词：物流管理 K-MEANS算法 遗传算法 收敛速度 云模型 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] TP311.13[自动化与计算机技术—控制科学与工程]