PageRank算法对页面评价太过客观,对不同重要程度的网页被授予相同的权重,并且在排序时,一些旧的页面经常出现在Web检索结果的前面,而新加入的高质量页面用户很难找到。针对Pagerank算法存在的这些缺陷,引入时间维加权概念,开发出TimedWPR算法,同时保证了两种页面的排序优化。该算法采用服务器反馈回来的网页修改时间表示网页年龄,并在此基础上对网络的组织结构和链接质量以及时间序列进行挖掘,从而克服现有Web超链接分析中的不足。
There are some limitation in allusion to the famous ranking algorithm Pagerank, such as the pages are evaluated too objectively, different pages with different importance is granted the same weight, some of the old pages always appear in the top, and some new pages get low ranks, etc. Therefore, the TimedWPR is proposed to solve these problems and to analyze the deficiency of nowadays link analysis. The algorithm uses the modification time returned by the web server as web age, and the web structure and time serious are analyzed through combining temporal dimension and link quality.