基于遗传聚类算法的Web日志挖掘研究  

Research on web log mining based on genetic clustering algorithm

作　　者：张艳肖[1] 

机构地区：[1]河北经贸大学计算机中心,石家庄050061

出　　处：《信息技术》2011年第12期10-12,16,共4页Information Technology

基　　金：河北省科技攻关计划项目(072135181)

摘　　要：K-均值聚类对初始聚类中心的选取较敏感,容易陷入局部最优。将改进的遗传算法与K-均值聚类相结合,以优化聚类中心。在种群进化过程中,父代个体均从种群中适应度高的个体中选择,同时,根据个体适应度动态调节交叉概率和变异概率,避免早熟现象。文中采用改进的遗传算法,对学院网站服务器上的Web日志进行用户和页面聚类,达到了很好的聚类效果。K-means clustering algorithm has the shortcoming that plunges into a local optimum prematurely because of sensitive selection of initial cluster center. Using improved genetic algorithm into K-means clustering algorithm can optimize the cluster centers. In the evolutionary process, the parent individuals that have high fitness are selected. At the same time, Pc and Pm are adjusted dynamic according to the individual fitness to avoid premature convergence. This paper researched on users cluster and pages cluster for college ＇ s web logs and a good cluster effect by using the improved genetic algorithm is gotten.

关 键 词：WEB日志挖掘 聚类 遗传算法 K-均值聚类 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]