一种利用Vague集理论改进的协同过滤推荐算法  

An Improved Collaborative Filtering Recommendation Algorithm Based on Vague Sets Theory

作　　者：张慧颖[1] 薛福亮[1,2] 

机构地区：[1]天津大学管理与经济学部,天津300072 [2]天津财经大学商学院,天津300222

出　　处：《现代图书情报技术》2012年第3期35-39,共5页New Technology of Library and Information Service

基　　金：天津市高等学校人文社会科学基金项目"电子商务个性化推荐系统瓶颈问题研究"(项目编号:20112125)的研究成果之一

摘　　要：针对项目特征难以表达问题,提出采用Vague集理论对其进行提取与表示,在此基础上进行项目相似性聚类,利用聚类内项目相似性对未评分项目进行评分值预测,从而消除协同过滤推荐的稀疏性问题,进而基于预测后的评分矩阵进行相似用户聚类,并在项目聚类簇内实施协同过滤推荐,使推荐更有针对性。实验结果表明,该方法无论在推荐精度还是推荐相关性上都更为有效。Aiming at the difficulty of project features expression,this paper brings forward to extract and represent it with vague sets theory. Then similar item is clustered to predict missing evaluation values of item, thus eliminating the sparsity problem of collaborative filtering recommendation. Based on the predicted rating matrix, similar users are clustered, and collaborative filtering recommendation is implemented in the space of item cluster to give more targeted recommendation. Evaluation results show that the proposed method is more effective both in the accuracy and in relevance of recommendations.

关 键 词：推荐系统 协同过滤 项目相似性 基于内容推荐 VAGUE SETS 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]