一种基于社交网络的非负矩阵分解算法  

A Non-negative Matrix Decomposition Algorithm based on Social Networks

作　　者：谢海迪 周云 李彤岩 XIE Haidi;ZHOU Yun;LI Tongyan(School of Communication Engineering,Chengdu University of Information Technology,Chengdu 610225,China;Unit 78111,Chengdu 610200,China)

机构地区：[1]成都信息工程大学通信工程学院,四川成都610225 [2]78111部队,四川成都610200

出　　处：《成都信息工程大学学报》2024年第1期50-55,共6页Journal of Chengdu University of Information Technology

摘　　要：基于社交网络的推荐算法主要是将用户社交关系和评分信息相结合,有效解决因缺乏评分数据而引起的冷启动问题。但基于社交网络的推荐算法只针对用户之间的相关性进行分析,事实上用户之间的关系水平也会对推荐结果产生一定程度的影响。因此提出一种基于社交网络的非负矩阵分解算法CTSVD。CTSVD算法通过用户的社交网络进行信任和不信任的亲密度计算,更新用户之间信任值和不信任值,校正社交关系对预测结果的影响。通过在实际数据集Epinions的实验,验证CTSVD方法的准确性,并能较好地解决传统的冷启动问题。The current social network-based recommendation algorithm mainly combines user social relations and scoring information,so as to effectively solve the cold start problem caused by the lack of scoring data.However,the current social network-based recommendation algorithms only analyze the correlation between users,which can have an impact on the recommendation results.Therefore,this paper proposes a non-negative matrix factorization algorithm CTSVD based on social network.The CTSVD algorithm calculates the intimacy of trust and distrust through the user's social network,updates the trust value and distrust value between users,and corrects the influence of social relations on the prediction results.The resuts of experiments using the actual dataset Epinions shows that the accuracy of the CTSVD method is verified,and the traditional cold start problem can be solved well.

关 键 词：推荐系统 社交网络 不信任关系 亲密度 可靠度 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]