摘要
通过对Web日志文件进行分析,提出了一种用混合遗传聚类对Web用户的行为进行分析的方法.混合遗传聚类是标准遗传算法和K-中心点算法的有机结合.实验证明,该方法是一个具有全局最优解的聚类方法,其结果明显优于标准遗传聚类方法.该算法能够有效地剔除噪音,得到很好的用户聚类和页面聚类的结果,为网站的管理者设计个性化的商务网站提供了有效的决策依据.
A mixed genetic clustering algorithm of web clients is presented based on the analysis of web log mining. It’s composed of standard genetic algorithm and K-Medoid clustering algorithm. The method is proved to be a dynamic clustering algorithm with global optimization and is superior to standard genetic algorithm. Web log mining used in this algorithm can filter out the noise and get good result of the clusters in users and pages. It also provides the effective decision-making for the web master to design individualistic web sites.
出处
《宁波大学学报（理工版）》
CAS
2005年第1期57-59,共3页
Journal of Ningbo University：Natural Science and Engineering Edition
基金
浙江省教育厅基金(20030485)资助
