摘要
如何对网上用户分类是W eb挖掘领域应用最多的任务之一,本文尝试将关联分类方法应用到W eb用户分类模式的挖掘.我们首先对服务器日志文件进行预处理,形成一个访问事务集;然后对该事务集进行数据挖掘,找出所有满足最小信任度和支持度的类别关联规则;最后,我们用这些类别关联规则去预测用户的兴趣.实验证明此方法是有效的.
Classification of web users is one of the most commonly used tasks in web mining, and the paper tries to apply the method of association classification to find classification patterns of web users. First we preprocess the web log to get a set of access transactions. Then we mine the set and get all frequent class association rules that satisfy the minimum confidence and support. At last, we predict interests of users using those rules. The experiment proves that the method is effective.
出处
《安徽大学学报(自然科学版)》
CAS
北大核心
2006年第2期17-20,共4页
Journal of Anhui University(Natural Science Edition)
关键词
WEB挖掘
关联规则
分类
Web mining
association role
classification