基于模糊模拟的加权偏爱浏览模式的挖掘  

Mining weighted preferred browsing patterns based on fuzzy simulation

作　　者：吴瑞[1] 

机构地区：[1]山西师范大学数学与计算机学院,山西临汾041004

出　　处：《计算机工程与应用》2007年第11期135-137,共3页Computer Engineering and Applications

基　　金：山西省自然科学基金(the Natural Science Foundation of Shanxi Province of China under Grant No.2006011039)。

摘　　要：每个网页由不同的专家给出语义上的重要性评估,这些语义评估再被刻画成相应的模糊语言变量,通过模糊模拟的方法,这些模糊语言变量被转化成表示网页重要性的权重。此外,简单地认为用户的访问频度反映了用户的访问兴趣是不准确的,因此在提出的加权支持度和偏爱度概念的基础上,从建立的包含了所有用户浏览信息的FLaAT(Frequent Link and Access Tree)上,挖掘用户偏爱的加权浏览模式。试验证明该算法是行之有效的。In this paper,importance evaluations of each web page are provided by different experts.These linguistic evaluations are characterized as corresponding fuzzy linguistic variables.By fuzzy simulation,the fuzzy linguistic variables are transformed as the weights of web pages.In addition,it is inaccurate that only visiting frequency discloses the user interest.Therefore,based on the concept of weighted support and preference,mining user preferred browsing patterns can be done from FLaAT （Frequent Link and Access Tree）,which stores all user browsing information.The algorithm is effective by an experiment.

关 键 词：WEB挖掘 模糊语言变量 用户浏览模式 模糊模拟 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论]