一种缓解数据稀疏性的协同过滤推荐系统  

A Collaborative Filtering Recommendation System for Data Sparsity

作　　者：刘旭东[1] 

机构地区：[1]烟台职业学院,山东烟台264670

出　　处：《烟台职业学院学报》2011年第4期57-60,共4页Journal of Yantai Vocational College

基　　金：国家科技支撑计划资助项目(2008BAH21B03)

摘　　要：协同过滤技术被成功地应用于个性化推荐系统中,但随着系统规模扩大,它面临着数据稀疏性、冷启动和可扩展性等问题。针对数据稀疏性问题,提出了一种基于专家系统的协同过滤个性化推荐模型,这种推荐模型能够很好地缓解数据稀疏性问题,在中小型电子商务推荐系统中具有良好的推荐效果和应用前景。Collaborative filtering is a very successful technology for building personalized recommendation system.However,as the system scale expands,there are some problems such as data sparsity,cold start and scalability.To solve the data sparsity,this paper proposes a kind of personalized recommendation model integrating the expert system and the collaborative filtering technology.It can relieve the data sparsity well and has a good recommendation effect and application prospect in small and medium-sized e-commerce recommendation system.

关 键 词：协同过滤 电子商务推荐系统 专家系统 

分 类 号：TP393.09[自动化与计算机技术—计算机应用技术]