基于Tf-Idf和网页链接的PageRank改进算法  

IMPROVED PAGERANK ALGORITHM BASED ON TF-IDF AND WEBPAGE LINK

作　　者：戴上平[1] 郑波荣[1] 

机构地区：[1]华中师范大学计算机科学学院,湖北武汉430079

出　　处：《计算机应用与软件》2013年第5期301-302,330,共3页Computer Applications and Software

摘　　要：提出基于Tf-Idf和网页链接对传统的PageRank算法不足之处进行改进。该算法不仅较好地解决了PageRank主题漂移问题,而且在查准率和查全率方面也有较大的提高。通过实验证明,该算法可以获得优于传统PageRank算法的查询结果集。This paper proposes to improve the inadequacy of traditional PageRank algorithm based on Tf-Idf and Webpage link.This improved algorithm solves the theme drift problem of PageRank,and also makes quite big progress in precision and recall.Through experiment we prove that the improved algorithm can acquire better query result set than the traditional PageRank algorithm.

关 键 词：PAGERANK 查全率 搜索引擎 网页连接 主题漂移 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]