Aiming at the PageRank algorithm ranking only rely on the link between the web structure,leads to problems of average distribution of webpage PR,topic drift,emphasis old pages,a kind of web page ranking algorithm based on multi feature fusion is proposed in this paper.By increasing the weight factor,on the basis of the indegree and the outdegree,according to the weight distribution of the PR value,avoid the average distribution of weight;Add the user feedback factor and the theme related factor,use of the click and keyword ratio,to solve the problem of topic drift;Add time factor,according to the same period the number of pages to be crawled and the existence of time is proportional to the relationship,to enhance the new page ranking,optimize the phenomenon of the old web pages.Verified by simulation experiment,the improved algorithm returns the page information in precision and correlation compared with the traditional PageRank algorithm is significantly improved. |