Enhancing Site Style Tree Construction Algorithm

摘要

The World Wide Web has become a global information service center with a vast amount of news, advertisements, product and service information, and disparate information from diversi-fied sources. However, only a small portion of information is truly relevant and useful to the users who are seeking information on specific topics. In this paper, common relations among nodes are taken into consideration when constructing site style tree, and a new node type is introduced. Experimental results show that the proposed algorithm has higher precision and recall.