Improve on frequent access path algorithm in web page personalized recommendation model

Web logs record actions and behaviors of users. By mining and analyzing these logs we can find users browsing and access patterns, and this is very important and useful to the web site optimization and recommender. This paper first analyses the association-rules-based personalized recommender model which is very popular in web site recommender system, points out the limitation of the frequent access path algorithm in this model, and then improves it. At last, the paper shows by the test results that the improved algorithm can advance the recommending quality.