Distributed Web log Mining Based Collaborative Filtering Recommendation Algorithm
For distributed large commercial mirror sites, this paper presents a distribuzed web log mining based collaborative filtering recommendation algorithm. Based on user-browsing preference, the user access matrixes of different mirror sites are constructed, then synthesized and standardized. Taking the standardized synthesized user access matrix as raw data, this paper proposes utilizing web page similarities to predict the rating for pages not having been rated, thus increasing the pages that have been jointly rated among users. This method could effectively solve the sparsity of user ratings in collaborative filtering and improve the accuracy of the calculation of the nearest neighbor of the target user. The experimental results show that this algorithm is applicable to the popular distributed web server clustering architecture, avoids the inaccuracy and complexity of web pages manual ratings, effectively solves the sparsity of user rating data of the traditional collaborative filtering algorithm and enhances the recommendation quality.
Distributed Web log mining Multi-Agent recommendation system collaborative filtering preference
Xun Wang Yun Ling Biwei Li
College of Computer & Information Engineering, Zhejiang Gongshang University Hangzhou 310035, China
国际会议
杭州
英文
1114-1118
2006-10-12（万方平台首次上网日期，不代表论文的发表时间）