XIE Haidi,ZHOU Yun,LI Tongyan.A Non-negative Matrix Decomposition Algorithm based on Social Networks[J].Journal of Chengdu University of Information Technology,2024,39(01):50-55.[doi:10.16836/j.cnki.jcuit.2024.01.009]
一种基于社交网络的非负矩阵分解算法
- Title:
- A Non-negative Matrix Decomposition Algorithm based on Social Networks
- 文章编号:
- 2096-1618(2024)01-0050-06
- Keywords:
- recommendation system; social networking; a relationship of mistrust; intimacy; reliability
- 分类号:
- TP391
- 文献标志码:
- A
- 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。
- Abstract:
- 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.
参考文献/References:
[1] Sahu A K,Dwivedi P J A I.User profile as a bridge in cross-domain recommender systems for sparsity reduction[J].Applied Intelligence,2019,49(7):2461-2481.
[2] 项亮.推荐系统实践[M].北京:人民邮电出版社,2012:197.
[3] Sardianos C,Ballas Papadatos G,Varlamis I.Optimizing parallel collaborative filtering approaches for improving recommendation systems performance[J].Information,2019,10(5):155.
[4] Koren Y.Factorization meets the neighborhood:a multifaceted collaborative filtering model[C].KDD ’08:Proceedings of the 14th ACM SIGKDD international conference on Knowledge discovery and data mining,2008:426-434.
[5] 杨志君,叶东毅.基于加权的不完备非负矩阵分解算法[J].计算机应用,2010,30(5):1280-1283.
[6] Cui Z,Xu X.Personalizedrecommendation system based on collaborative filtering for loTscenarios[J].IEEETransactions on Services Computing,2020,13(4):685-695.
[7] Ma H,Yang H,Lyu M R,et al.Sorec:social recommendation using probabilistic matrix factorization [C].CIKM ’08: Proceedings of the 17th ACM conference on Information and knowledge management,2008:931-940.
[8] Yang B, Lei Y, Liu J, et al. Social collaborative filtering by trust[J].transactions on pattern analysis,2016,39(8):1633-1647.
[9] Jamali M,ESTER M.A matrix factorization technique with trust propagation for recommendation in social networks[C].RecSys ’10: Proceedings of the fourth ACM conference on Recommender systems,2010:135-142.
[10] Guo G,Zhang J,Yorke-Smith N.TrustSVD:Collaborative filtering with both the explicit and implicit influence of user trust and of item ratings[C].Twenty-Ninth AAAIConference on Artificial Intelligence,2015.
[11] 张紫茵,张恒汝,徐媛媛,等.用户非对称信任关系的推荐算法[J].计算机科学,2018,45(10):37-42.
[12] Zhang S,Zhu J.Reliable Potential FriendsIdentification Based on Trust Circuit for Social Recommendation[C].WASA 2020:Wireless Algorithms,Systems,and Applications,2020:716-729.
[13] 张琦,柳玲,文俊浩. 一种基于领域信任及不信任的奇异值分解推荐算法[J].计算机科学,2019,46(10):27-31.
[14] 郑鹏,王应明,梁薇.基于信任和矩阵分解的协同过滤推荐算法[J].计算机工程与应用,2018,54(13):34-40.
[15] 刘智捷. 基于融合信任关系的协同过滤推荐算法[J].杭州电子科技大学学报(自然科学版),2018,38(3).
[16] Resnick P,Zeckhauser R.Trust among strangers in Internet transactions:Empirical analysis of eBay’s reputation system[M].The Economics of the Internet and E-commerce,Emerald Group Publishing Limited,2002.
[17] 张大鹏,张伟.基于综合信任的奇异值分解推荐算法研究[J].高技术通讯,2021,31(1):102-112.
[18] 乔猛,魏国亮,吴超异.一种基于信任和不信任的矩阵分解推荐算法[J].小型微型计算机系统,2021,44(1):56-62.
相似文献/References:
[1]欧如月,陶宏才.一种融合PageRank和PersonalRank的多层个性化推荐算法[J].成都信息工程大学学报,2021,36(03):305.[doi:10.16836/j.cnki.jcuit.2021.03.011]
OU Ruyue,TAO Hongcai.A Multi-layer Personalized Recommendation Algorithm Combining PageRank and PersonalRank[J].Journal of Chengdu University of Information Technology,2021,36(01):305.[doi:10.16836/j.cnki.jcuit.2021.03.011]
备注/Memo
收稿日期:2022-10-18