Study on Multi-criteria Personalized Recommendation Algorithm

摘要：

　　Personalized recommendation plays an important role in E-commerce.Single-criteria recommendation algorithm has the features of simplicity and high efficiency but also can easily have the problems of cold startup and sparse data.Multi-criteria recommendation algorithm is to carry out prediction from two prospects:customer and commodity,and fully evaluate and give rates to customer similarity,customers evaluation for commodity,rank of commodity sales,and commodity similarity,etc.The commodities in a same layer are given an aggregated rate respectively by borrowing the concept of commodity hierarchical tree from aggregation.Eventually the advantages of multi-criteria recommendation algorithm are summarized.

关键词： E-commerce Personalized Recommendation Multi-criteria Algorithm Aggregate Function

作者: Luo Ya Zhao Li

作者单位: Guizhou University of Finance and Economics

会议类型: 国际会议

会议名称: 2012 2nd International Conference on Computer Application and System Modeling(2012第二届计算机应用与系统建模国际会议)(ICCASM-2012)

会议地点: 沈阳

会议语种:英文

页码: 129-131

在线出版日期: 2012-07-27（万方平台首次上网日期，不代表论文的发表时间）