基于PageRank的文献重要度评价方法

Most of the traditional citation analysis methods only use the citation counts and the factor of journal to evaluate documents, and existing papers visualization layout algorithms cannot reflect the importance of papers. We propose a new evaluating method based on the PageRank algorithm and build its two-layer visualization layout model. Firstly, this method improves the existing PageRank algorithm via considering the behavior of citation, the authority of author and time effects on the value of documents. Then it evaluates the value of documents by analyzing the paper's inherent value and the value passed from the behavior of citation. Lastly, we build a two-layer visualization layout model based on citation and co-author network, intuitively visualize evaluated results. This method has been tested by using CNKI dataset. Experiments show that it successfully solves the problem that the results tend to the early papers and topic drift, and can show the results more clearly.