基于矩阵聚类的协作过滤算法  

Matrix clustering-based collaborative filtering algorithm

作　　者：高凤荣[1] 邢春晓[1] 杜小勇[2] 王珊[2] 

机构地区：[1]清华大学信息技术研究院,北京100084 [2]中国人民大学信息学院,北京100872

出　　处：《华中科技大学学报（自然科学版）》2005年第z1期257-260,共4页Journal of Huazhong University of Science and Technology(Natural Science Edition)

基　　金：国家自然科学基金资助项目(60473078)

摘　　要：通过分析现有的协作过滤技术,提出了基于矩阵聚类的协作过滤算法,把矩阵聚类算法和协作过滤相结合,自动划分原始用户———资源评分矩阵,依据划分后的子数据矩阵生成推荐结果.实验结果表明,提出的基于矩阵聚类的协作过滤算法优于传统协作过滤算法,减少了近邻搜索范围,提高了算法的推荐精度.Collaborative filtering is the most successful personalized service technique to date and is used in many domains.Unfortunately collaborative filtering is limited by the huge dimensionality and sparsity of user-item rating matrix.In this paper,matrix clustering-based collaborative filtering method is proposed to overcome the drawback of collaborative filtering algorithms.The user-item rating matrix can be partitioned into low-dimensional dense matrices using matrix clustering algorithm.The recommendations are generated based on low-dimensional data matrices.An extensive performance study shows that the proposed approach outperforms traditional collaborative filtering algorithm.It can reduce neighbor search space and improve the accuracy of recommendations.

关 键 词：协作过滤 聚类分析 稀疏矩阵 个性化服务 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]