摘要
Web日志挖掘的一个主要任务是获得用户的浏览模式,这对Web站点的改进和为用户提供个性化服务提供了有价值的潜在信息。本文提出了一种改进算法——RD_Close。该算法对数据挖掘中的Close方法进行了改进,并引入了RD_Apriori方法中缩减数据库的思想。这种改进算法能有效发现用户频繁访问模式。最后,用实际数据对算法性能进行了充分验证和深入分析。
Discovering and identifying users' access patterns, which can provide very valuable potential information for the improvement of web site and the personalized service of users , is one of the primary tasks on web usage mining. An improved algorithm-RD_Close is proposed, which has improved the Close method on data mining and introduced the thought of reduced database in RD _Apriori method. The improved algorithm can discover user frequent access patterns effectively. The performance of algorithm has been adequately tested and deeply analyzed by actual data.
出处
《模式识别与人工智能》
EI
CSCD
北大核心
2005年第6期728-734,共7页
Pattern Recognition and Artificial Intelligence
基金
安徽省自然科学基金(No.050420207)
合肥工业大学科研发展基金(No.030503F)
关键词
WEB挖掘
频繁访问模式
访问模式的中心交集
访问模式的中心子集
封闭访问模式
Web Mining, Frequent Access Pattern, Central Intersection of Access PatternsCentral Subset of Access Patterns, Closed Access Pattern