Social Recommendation Model Based on Self-Supervised Tri-Training and Consistent Neighbor Aggregation
LIU Shudong1,2, LI Liying1,2, CHEN Xu1,2
1. Centre for Artificial Intelligence and Applied Research, Zhongnan University of Economics and Law, Wuhan 430073; 2. School of Information and Engineering, Zhongnan University of Economics and Law, Wuhan 430073
摘要 将用户社会关系引入用户-商品评分数据中,构建用户-商品异构关系图,可缓解传统推荐系统面临的数据稀疏性和冷启动问题.但是,由于用户间社会关系的复杂性,聚合不一致的社会邻居可能会降低推荐性能.针对上述问题,文中提出基于自监督三重训练和聚合一致邻居的社会化推荐模型(Social Recommendation Based on Self-Supervised Tri-Training and Consistent Neighbor Aggregation, SR-STCNA).首先,在用户-商品评分数据的基础上,引入用户-用户间的社交关系,在用户-商品异构图中构建多种关系.使用超图表示用户和用户、用户和商品之间的关系.使用自监督三重训练,从未标记的数据中学习用户表示,充分挖掘用户-用户和用户-商品间存在的复杂连接关系.然后,通过用户-商品异构图上的节点一致性得分和关系自注意力,在用户和商品表示学习过程中聚合一致邻居,增强用户和商品嵌入表示能力,提高推荐性能.在CiaoDVD、FilmTrust、Last.fm、Yelp数据集上的实验表明,SR-STCNA性能较优.
Abstract:Integrating user social relationships into user-item rating data to construct a heterogeneous user-item graph can alleviate data sparsity and cold start in traditional recommender systems. However, due to the complexity of user social relationships, aggregating inconsistent neighbors may degrade the recommendation performance. To address this issue, a social recommendation model based on self-supervised tri-training and consistent neighbor aggregation(SR-STCNA) is proposed. Firstly, on the basis of user-item rating data, social relationships among users are introduced and diverse relations within the heterogeneous user-item graph are established. The relationships between users as well as between users and items are presented by a hypergraph. Self-supervised tri-training is employed to learn users' representations from unlabeled data and uncover the complex connectivity between user-user and user-item interactions. Then, the consistent neighbors of users and items are aggregated in the process of their representation learning by the node consistency score and relationship self-attention on the user-item heterogeneous graph. Consequently, the representation ability of users and items is enhanced, thereby improving the recommendation performance. Finally, the experimental results on CiaoDVD, FilmTrust, Last.fm and Yelp datasets validate the superiority of SR-STCNA.
刘树栋, 李丽颖, 陈旭. 基于自监督三重训练和聚合一致邻居的社会化推荐模型[J]. 模式识别与人工智能, 2024, 37(3): 207-220.
LIU Shudong, LI Liying, CHEN Xu. Social Recommendation Model Based on Self-Supervised Tri-Training and Consistent Neighbor Aggregation. Pattern Recognition and Artificial Intelligence, 2024, 37(3): 207-220.
[1] KOREN Y, BELL R, VOLINSKY C.Matrix Factorization Techni-ques for Recommender Systems. Computer, 2009, 42(8): 30-37. [2] SHAN Y, HOENS T, JIAO J, et al. Deep Crossing: Web-Scale Modeling Without Manually Crafted Combinatorial Features // Proc of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York,USA: ACM, 2016: 255-262. [3] HE X N, LIAO L Z, ZHANG H W, et al. Neural Collaborative Filtering // Proc of the 26th International Conference on World Wide Web. New York,USA: ACM, 2017: 173-182. [4] CHENG H T, KOC L, HARMSEN J, et al. Wide & Deep Learning for Recommender Systems // Proc of the 1st Workshop on Deep Learning for Recommender Systems. New York,USA: ACM, 2016: 7-10. [5] RAFAILIDIS D, NANOPOULOS A.Modeling Users Preference Dynamics and Side Information in Recommender Systems. IEEE Transactions on Systems, Man, and Cybernetics(Systems), 2016,46(6): 782-792. [6] CHRISTAKOPOULOU K, RADLINSKI F, HOFMANN K.Towards Conversational Recommender Systems // Proc of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York,USA: ACM, 2016: 815-824. [7] 孟祥武,刘树栋,张玉洁,等.社会化推荐系统研究.软件学报, 2015, 26(6): 1356-1372. (MENG X W, LIU S D, ZHANG Y J, et al. Research on Social Recommender Systems. Journal of Software, 2015, 26(6): 1356-1372.) [8] KING I, LYU M R, MA H.Introduction to Social Recommendation // Proc of the 19th International Conference on World Wide Web. New York,USA: ACM, 2010: 1355-1356. [9] JIANG M, CUI P, LIU R, et al. Social Contextual Recommendation // Proc of the 21st ACM International Conference on Information and Knowledge Management. New York,USA: ACM, 2012: 45-54. [10] TANG J L, WANG S H, HU X, et al. Recommendation with Social Dimensions. Proceedings of the AAAI Conference on Artificial Intelligence, 2016, 30(1): 251-257. [11] YU J L, GAI M, LI J D, et al. Adaptive Implicit Friends Identification over Heterogeneous Network for Social Recommendation // Proc of the 27th ACM International Conference on Information and Knowledge Management. New York,USA: ACM, 2018: 357-366. [12] CHEN C, ZHANG M, WANG C Y, et al. An Efficient Adaptive Transfer Neural Network for Social-Aware Recommendation // Proc of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval. New York,USA: ACM, 2019: 225-234. [13] WU L, SUN P J, HONG R C, et al. Collaborative Neural Social Recommendation. IEEE Transactions on Systems, Man, and Cybernetics(Systems), 2021, 51(1): 464-476. [14] FAN W Q, MA Y, LI Q, et al. Graph Neural Networks for Social Recommendation // Proc of the World Wide Web Conference. New York,USA: ACM, 2019: 417-426. [15] SONG C H, WANG B, JIANG Q X, et al. Social Recommendation with Implicit Social Influence // Proc of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval. New York,USA: ACM, 2021: 1788-1792. [16] HUANG C, XU H C, XU Y, et al. Knowledge-Aware Coupled Graph Neural Network for Social Recommendation. Proceedings of the AAAI Conference on Artificial Intelligence, 2021, 35(1): 4115-4122. [17] LIU Y, CHEN L, HE X N, et al. Modelling High-Order Social Relations for Item Recommendation. IEEE Transactions on Know-ledge and Data Engineering, 2022, 34(9): 4385-4397. [18] YU J L, GAO M, YIN H Z, et al. Generating Reliable Friends via Adversarial Training to Improve Social Recommendation // Proc of the IEEE International Conference on Data Mining. Washington, USA: IEEE, 2019: 768-777. [19] QUAN Y H, DING J J, GAO C, et al. Robust Preference-Guided Denoising for Graph Based Social Recommendation // Proc of the ACM Web Conference. New York,USA: ACM, 2023:1097-1108. [20] FU B R, ZHANG W M, HU G N, et al. Dual Side Deep Context-Aware Modulation for Social Recommendation // Proc of the Web Conference. New York,USA: ACM, 2021:2524-2534. [21] YU J L, YIN H Z, LI J D, et al. Self-Supervised Multi-channel Hypergraph Convolutional Network for Social Recommendation // Proc of the ACM Web Conference. New York,USA: ACM, 2021: 413-424. [22] KHAN B, WU J, YANG J, et al. Heterogeneous Hyper-Graph Neural Network for Social Recommendation Using Attention Network. ACM Transactions on Recommender Systems, 2023. DOI: 10.1145/3613964. [23] WU B C, KANG Y Y X, GUAN B, et al. We Are Not So Similar: Alleviating User Representation Collapse in Social Recommendation // Proc of the ACM International Conference on Multimedia Retrieval. New York,USA: ACM, 2023: 378-387. [24] SUN Y C.Denoising Explicit Social Signals for Robust Recommendation // Proc of the 17th ACM Conference on Recommender Systems. New York,USA: ACM, 2023: 1344-1348. [25] BRETTO A, CHERIFI H, ADOUTAJDINE D.Hypergraph Imaging: An Overview. Pattern Recognition, 2002, 35(3): 651-658. [26] YU J L, YIN H Z, GAO M, et al. Socially-Aware Self-Supervised Tri-Training for Recommendation // Proc of the 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. New York,USA: ACM, 2021: 2084-2092. [27] ERICSSON L, GOUK H, LOY C C, HOSPEDALES TM.Self-Su-pervised Representation Learning: Introduction, Advances and Cha-llenges. IEEE Signal Processing Magazine, 2022, 39(3): 42-62. [28] YU L, PEI S C, DING L Z, et al. SAIL: Self-Augmented Graph Contrastive Learning. Proceedings of the AAAI Conference on Artificial Intelligence, 2022, 36(3): 8927-8935. [29] HE X N, DENG K, WANG X, et al. LightGCN: Simplifying and Powering Graph Convolution Network for Recommendation // Proc of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval. New York,USA: ACM, 2020: 639-648. [30] VAN DER OORD A, LI Y Z, VINYALS O. Representation Lear-ning with Contrastive Predictive Coding[C/OL].[2023-11-09]. https://arxiv.org/pdf/1807.03748.pdf. [31] RENDLE S, FREUDENTHALER C, GANTNER Z.BPR: Baye-sian Personalized Ranking from Implicit Feedback // Proc of the 25th Conference on Uncertainty in Artificial Intelligence. Arlington, USA: AUAI press, 2009: 452-461. [32] WU L, LI J W, SUN P J, et al. DiffNet++: A Neural Influence and Interest Diffusion Network for Social Recommendation. IEEE Transactions on Knowledge and Data Engineering, 2020, 34(10): 4753-4766. [33] YU J L, YIN H Z, XIA X, et al. Are Graph Augmentations Ne-cessary?: Simple Graph Contrastive Learning for Recommendation // Proc of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval. New York,USA: ACM, 2022: 1294-1303.