Using Fuzzy Semantic Log for Rough Set Web Page Recommendation

Improving accuracy of Web page recommendation through data mining technology is an important research topic. This paper presents a new rough set Web page recommendation algorithm based on fuzzy semantic logs, which firstly changes the access logs in the Web into fuzzy semantic logs, secondly matches the current session with the rules founded, finally gives a recommendation set of web pages to the users. To evaluate the effectiveness of the algorithm the backward path ratio method is used, and the result shows that the algorithm can effectively improve the accuracy of Web page recommendation.