基于兴趣度的聚类协同过滤推荐系统的设计

协同过滤技术被成功地应用于个性化推荐系统中.随着用户数目和网页数目的日益增加,整个用户矩阵数据极端稀疏并且实时性效果不理想.传统的推荐方法解决不了这些问题.本文结合兴趣度和聚类技术对客户的个人兴趣进行评价,提出了基于兴趣度的聚类协同过滤推荐系统,实验表明,该算法能够有效避免传统方法带来的弊端,提高系统的推荐质量.

An clustering and optimized collaborative filtering recommendation based on interest measure

Collaborative filtering is used extensively in personalized recommendation systems.With the magnitudes of users and commodities grow rapidly,resulting in the extreme sparsity of user rating data and the decreasing of real time performance.Traditional recommendation system work poor in this situation.This paper integrated collaborative filtering into interest measure to analyze the customer's personal taste,presents an clustering and optimized collaborative filtering recommendation based on interest measure.It can be proved that the new algorithm presented has a better performance corresponding to the known algorithm.The experiment shows that the approach is successful.