Abstract: Aiming at PageRank algorithm used by Google, this paper proposes an improved Web mining algorithm. When realizing the algorithm, Web page using information and adding date information turn to click vector and date vector. It receives a vector after two vectors weighting of standardization. Then the vector will be put into the process of the improved iterative algorithm as a constant. Experimental result proves that the improved algorithm can make a good accuracy on evaluating the importance of Web page.

Key words: search engine, Web page, PageRank algorithm

摘要： 针对Google使用的PageRank算法，提出一种改进的Web挖掘算法。实现该算法时，将网页使用信息和网页添加日期信息做成点击向量和日期向量，2个向量加权后标准化得到的一个向量作为常数加入到改进的迭代算法。实验结果证明，改进算法可以提高网页重要性判断的准确度。

关键词: 搜索引擎, 网页, PageRank算法

CLC Number: 

• TP391