一种改进的PageRank算法

针对PageRank算法存在主题漂移以及偏重旧网页的问题,结合锚文本相似度和时间反馈因子提出了一种PageRank改进算法STPR,并对STPR算法进行实验分析。先比较了传统PageRank算法与加入锚文本相似度的PageR-ank算法,结果表明加入锚文本相似度的PageRank算法有利于减少主题漂移现象的发生;其次比较了加入锚文本相似度的PageRank算法与STPR算法,结果表明STPR算法不但减少了主题漂移现象,而且还弥补了新网页的PageRank值。

An Improved PageRank Algorithm

Aiming at the problems of Topic-drift and emphasizing on old web pages for PageRank algorithm,combined with anchor texts similarity and timing feedback factor,this article presents an improved algorithm STPR,and analyzs STPR algorithm by experiment.First this article compares the traditional PageRank algorithm and adding anchor text similarity PageRank algorithm,results show that by adding anchor text similarity PageRank algorithm helps to reduce the occurrence of the phenomenon of topic-drift.Second this article compares adding anchor text similarity PageRank algorithm and STPR algorithm,results show that STPR algorithm not only reduce topic-drift phenomenon,but also make up PageRank value for new web pages.