基于最大流及页面相似度的Web结构挖掘  

Web Structure Mining Based on Maximum Flow and Page Similar Value

作　　者：李莹[1] 吴晓军[1] 

机构地区：[1]陕西师范大学计算机科学学院,陕西西安710062

出　　处：《计算机技术与发展》2011年第10期112-115,共4页Computer Technology and Development

基　　金：中央高校基本科研业务费专项资金资助(GK201002005);陕西省工业攻关计划(2009K09-21)

摘　　要：针对Web结构挖掘算法容易出现"主题漂移"以及主机间的多重互相加强关系的问题,提出了一种基于最大流与页面相似度值的超链接结构挖掘方法。该方法在传统的超链接结构挖掘算法HITS的基础上引入页面相似度值构造邻接矩阵,并结合基于最大流的Web社区发现技术来构建特征向量空间模型,通过迭代计算最终获得价值最高的权威结果集和中心结果集。实验结果证明该方法有较好的查准率与查全率,并有效抑制了"主题漂移"现象,具有一定的实用价值。Aiming to Web structure mining algorithm is easy for a ＂topic drift＂ and mutually strengthening relations among the hots of the problem, a method of hyperlink structure mining based on the maximum flow and the page similarity value is presented. On the basis of traditional HITS algorithm, this method introduced the page similarity value and adopted the Web communities identification based on the maximum flow to construct the models of feature vector space. And then the calculation eventually won the highest Nalue of authority-set and hub-set by iterative method. Experimental results show that the method has better recall and precision, what＇ s more it effectively inhibits the theme of Web structure mining algorithms drift, has some practical value.

关 键 词：WEB结构挖掘 主题漂移 页面相似度值 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]