SigRA:A New Similarity Computation Method in Recommendation System

摘要：

　　Because the collaborative filtering(CF)method has many advantages,it is widely used in recommendation systems.Measuring the similarity between users or items is an important step in collaborative filtering.The accuracy of existing similarity calculation methods should be improved.In this paper,we propose a new method for similarity calculation,which is called SigRA.This new method considers both the users ratings of the items and the rated items by the users.Experiments on the MovieLens data set have been conducted using the proposed method,and the results are compared with that of several existing methods to verify the validity of the proposed method.Experimental results show that the proposed method is superior to the compared methods in terms of accuracy.

作者: Xiaokun Wu Yongfeng Huang

作者单位: Tsinghua National Laboratory for Information Science and Technology Department of Electronic Engineering,Tsinghua University Beijing,100084,China

会议类型: 国际会议

会议名称: 第九届网络分布式计算与知识发现国际会议( 2017 International Conference on Cyber-enabled distributed computing and knowledge discovery)

会议地点: 南京

会议语种:英文

页码: 148-154

在线出版日期: 2017-10-12（万方平台首次上网日期，不代表论文的发表时间）