摘要
介绍了提高Web访问效率必要性、数据挖掘的概念、聚类技术的原理、及其实现的步骤。聚类分析技术是根据数据对象的属性信息或对象间的关系,将数据对象分成类簇(Cluster),同一簇中的对象有较高的相似度,不同簇中的对象彼此差别较大。利用聚类技术可以发现一同被访问的Web页面,并把它们编到一个组里,帮助用户更好地访问,提高Web访问效率。可以通过对服务器的访问日志进行数据挖掘,得出网站页面之间共同被访问的频度,使用数据矩阵寻找每个聚类,根据每个聚类创立索引页,通过索引页帮助用户访问,提高访问速度。
Clustering is divided data into class or cluster according to the nature of data or relationship between data. In the same cluster, there is high similarity, while is different cluster, differences exist. The interviewed web will be found and put into group by clustering technology. Which help users interview and improve efficiency. Data mining in interview Server Log data, the interview frequency are calculated. Then using data matrix and looking for each cluster in the data matrix, an index is set up to help user＇s interview web and promote the interview speed.
出处
《机械管理开发》
2006年第4期5-6,8,共3页
Mechanical Management and Development
关键词
数据挖掘
聚类技术
数据矩阵
访问日志
Date mining Clustering technology Data matrix Access log
