摘要
提出了一种Web用户兴趣变化识别的算法,面向层次结构的信息类网站,可以达到客户自适应的目的.该算法由长、短期两个兴趣模型构成,分别基于指数衰减理论和贝叶斯后验概率理论.基于一个Internet上真实网站的实验结果表明,本文提出的算法可以迅速识别出用户的兴趣变化,并且广泛适用于新闻、虚拟社区等层次结构类网站的网页内容推荐,满足用户在信息浏览时的个性化需求.
To identify Web users' interest shift on hierarchical structure websitcs, this paper present a new algorithm, which is used for adaptive website building and composed of two models based on exponential decay and Bayesian posterior probability respectively. Experiments based on a real Internet website show that the proposed algorithm can identify the user interests' shift promptly, as well as satisfying personalized information demands. The proposed algorithm can thus be widely implemented on various websites such as online News and virtual communities.
出处
《系统工程理论与实践》
EI
CSCD
北大核心
2008年第10期89-95,共7页
Systems Engineering-Theory & Practice
基金
国家自然科学基金(70401010)
国家留学基金