作　　者： ; ;

机构地区： 中山大学信息科学与技术学院计算机科学系

出　　处： 《计算机工程与应用》 2005年第18期162-165,共4页

摘　　要： 随着Internet的基础结构不断扩大和其所含信息的持续增长,Internet用户越来越感觉容易在WWW服务中“资源迷向”。提高用户访问效率的方法有页面预取技术,站点动态重构技术和web个性化推荐技术等。现有的大多数web个性化推荐技术主要是基于用户使用记录的数据挖掘方法,没有或很少考虑结合页面内容—这才是用户真正感兴趣的。该文提出一种结合用户使用挖掘和内容挖掘的web推荐服务,该推荐服务根据频繁最大前向访问路径,提出含有导航页和内容页的频繁访问路径图概念,根据滑动窗口内的最近用户访问页面内容和候选推荐集中页面内容相关性,来向用户提供个性化推荐服务。经推荐质量分析,这种方法具有较好的推荐优化能力。 As the basic construction of the Internet constantly expands and the information it contains continuously increases,Internet users find being apt to get lost in resource searching in WWW service.The ways in which the efficiency of user browsing can be improved are page prefetching,dynamic website reconstruction and web personalized recommendation etc.Most of the existing web personalized recommendation are mainly data mining methods based on the using record,with few or no regard to the content of the page,which is actually the users' interest.This paper puts forward a web recommendation service that takes into account both web usage mining and web content mining.Based on the frequent max forward path,it raises the concept of frequently visited path diagram including the navigation page and content page.According to the relevancy of the content of the latest visited page in the sliding window and that of the candidate recommended page,this service provides users with the personalized recommendation pages.As it is proved by the recommendation quality analysis,this approach has a good performance of optimal recommendation.

关 键 词： 个性推荐 使用挖掘 内容挖掘

领　　域： [自动化与计算机技术] [自动化与计算机技术]