摘要
针对评分数据稀疏的情况下传统相似性计算的不足,提出了一种基于项目之间相似性的协同过滤算法。该算法结合用户对项目的评分和项目之间的兴趣度进行项目之间的相似性计算,在一定程度上减小了评分数据稀疏的负面影响。实验结果表明,该算法在评分数据稀疏的情况下,能使推荐系统的推荐质量明显提高。
To solve the problems of traditional similarity measure methods with user rating data sparsity,this paper proposed a novel collaborative filtering algorithm based on item similarity,which combined user rating data with interest degree of items to calculate similarity between two items,so that it could overcome the effect of sparsity of user rating data.The experimental results show that the proposed algorithm can obviously enhance the quality of recommendation system in the case of sparsity of user rating data.
出处
《计算机应用研究》
CSCD
北大核心
2012年第1期116-118,126,共4页
Application Research of Computers
基金
重庆市科委科技项目(CSTC2009CB2015)
中韩国际合作项目(C2010-02)
关键词
兴趣点
推荐系统
协同过滤
相似性
项目兴趣度
point of interest(POI)
recommendation system
collaborative filtering
similarity
item interest degree