基于网页结构相关性的隐马尔可夫预取技术研究  

Research on Structural-Relation-based Hidden Markov Model Web Pre-fetching

作　　者：张泊平[1] 张得喜[1] 

机构地区：[1]许昌学院计算机科学与技术学院,许昌461000

出　　处：《计算机与数字工程》2007年第5期88-90,102,共4页Computer & Digital Engineering

基　　金：国家自然科学基金项目(编号:60475040)资助;河南省科技攻关项目(编号:0524220054)资助

摘　　要：基于网页的结构相关性及隐马尔可夫过程提出一种基于Web网页结构相关性的预取模型,通过网页抓捕建立特征词集,在此特征词集的基础上对用户的访问序列进行分析,提取超链接语义中蕴含的信息需求,在此特征词集的基础上对网页超链语义进行分析。模型引入隐马尔可夫模型实现用户访问序列中潜在意图的挖掘。性能测试实验的结果表明,该模型具有较好的整体性能。A Pre-fetching model is brought forward based on Web pages structural-relation and Hidden Markov process. The Pre-fetching Model picks-up the semantic cues of links contained in the user’s information request through analyzing the user’s visiting list, and snatches at a great deal of pages by way of semantic cue from different pages using network crawl software, and analyzes the semantic cues of links base on the character words, then computes the probability of the user’s visiting list. HMM is imported to achieve mining latency intents of the user’s visiting list. The capability test make it cheat that it has a preferably capability, at the same time it assures pre-fetching veracity it has preferably usability and high hit the target rate and high speed, it can minish the delay of the user’s visiting, and improve response speed.

关 键 词：WEB预取 结构相关性 隐马尔可夫模型 超链 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]