Multi-View Contrastive Learning for Recommendation with Meta-Knowledge and SVD

doi:10.3778/j.issn.1002-8331.2408-0107

Abstract

Abstract: Current heterogeneous graph contrastive learning typically relies on a single method to construct auxiliary views for data augmentation. Considering the complexity of interactions between users and items under heterogeneous modeling, single-method approaches often fall short in capturing the rich semantic information inherent in user-item interactions. To address this, this paper proposes a multi-view contrastive learning (MVCL) recommendation model with meta-knowledge and SVD (singular value decomposition). This model introduces a personalized data augmentation module that constructs auxiliary views from multiple perspectives. Specifically, the construction methods of auxiliary views include two key points： Firstly, for intrinsic relationships between users and items, auxiliary views are constructed through personalized mapping via meta-knowledge fusion; Secondly, for user-item interactions, auxiliary views are constructed by using approximate singular value decomposition to achieve adaptive contrastive enhancement. Additionally, an enhanced robust contrastive loss function is utilized to mitigate noise sensitivity, further improving recommendation performance. Experiments on the CiaoDVD, Epinions, and Yelp datasets demonstrate that MVCL consistently outperforms existing models, validating its effectiveness in addressing data sparsity and capturing rich semantic information in heterogeneous contexts.

Key words: recommendation system, heterogeneous graph, contrastive learning, auxiliary views, meta-knowledge learning

摘要： 目前的异构图对比学习通常依赖于单一的方法来构建辅助视图以进行数据增强。考虑到异构建模下用户和商品间交互的复杂性，单一构造方法下的辅助视图往往难以有效捕捉异构数据中丰富的语义信息。为此，提出了融合元知识和奇异值分解（singular value decomposition，SVD）的多视图对比学习推荐模型（multi-view contrastive learning，MVCL）。引入了个性化数据增强模块，并采用不同方法从多角度构建了辅助视图。其中辅助视图的构建方法具体包括两点，分别是：针对用户商品间的内在联系，通过元知识融合来个性化映射构造辅助视图；针对用户商品间交互，通过近似奇异值分解的方法构建辅助视图，实现自适应对比增强。为有效解决传统对比损失抗噪声能力较弱的问题，使用了鲁棒性更强的增强型对比损失函数作为输出，进一步提升了模型的推荐性能。在此基础上，为进一步验证所提出模型的有效性，在CiaoDVD、Epinions、Yelp三个公开数据集上进行了实验，实验结果表明提出的新模型在不同场景下均呈现了较好的性能，有效验证了该模型在解决异构数据稀疏性和利用丰富语义信息方面的可行性。

关键词: 推荐系统, 异构图, 对比学习, 辅助视图, 元知识学习

LIU Jingxiang, WANG Feng, WEI Wei. Multi-View Contrastive Learning for Recommendation with Meta-Knowledge and SVD[J]. Computer Engineering and Applications, 2025, 61(22): 159-169.

刘景祥, 王锋, 魏巍. 融合元知识和SVD的多视图对比学习推荐系统[J]. 计算机工程与应用, 2025, 61(22): 159-169.

References

[1] XIONG H T, LIU Z B. A situation information integrated personalized travel package recommendation approach based on TD-LDA model[C]//Proceedings of the 2015 International Conference on Behavioral, Economic and Socio-cultural Computing. Piscataway: IEEE, 2015: 32-37.
[2] WANG S F, GONG M G, WU Y, et al. Multi-objective optimization for location-based and preferences-aware recommendation[J]. Information Sciences, 2020, 513: 614-626.
[3] WU S W, SUN F, ZHANG W T, et al. Graph neural networks in recommender systems: a survey[J]. ACM Computing Surveys, 2023, 55(5): 1-37.
[4] CHEN L, WU L, HONG R C, et al. Revisiting graph based collaborative filtering: a linear residual graph convolutional network approach[J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2020, 34(1): 27-34.
[5] GU S Y, WANG X, SHI C, et al. Self-supervised graph neural networks for multi-behavior recommendation[C]//Proceedings of the 31st International Joint Conference on Artificial Intelligence, 2022: 2052-2058.
[6] LIU W M, SU J J, CHEN C C, et al. Leveraging distribution alignment via stein path for cross-domain cold-start recommendation[C]//Advances in Neural Information Processing Systems 34, 2021: 19223-19234.
[7] 黄若然, 崔莉, 韩传奇. 推荐系统中稀疏情景预测的特征-类别交互因子分解机[J]. 计算机研究与发展, 2022, 59(7): 1553-1568.
HUANG R R, CUI L, HAN C Q. Feature-over-field interaction factorization machine for sparse contextualized prediction in recommender systems[J]. Journal of Computer Research and Development, 2022, 59(7): 1553-1568.
[8] ZHANG Q R, XIA L H, CAI X H, et al. Graph augmentation for recommendation[C]//Proceedings of the 2024 IEEE 40th International Conference on Data Engineering. Piscataway: IEEE, 2024: 557-569.
[9] YOU Y N, CHEN T L, SUI Y D, et al. Graph contrastive learning with augmentations[C]//Advances in Neural Information Processing Systems 33, 2020: 5812-5823.
[10] 吴国栋, 吴贞畅, 王雪妮, 等. 图对比学习研究进展[J]. 小型微型计算机系统, 2025, 46(1): 44-54.
WU G D, WU Z C, WANG X N, et al. Graph contrastive learning research progress[J]. Journal of Chinese Computer Systems, 2025, 46(1): 44-54.
[11] CHEN M R, HUANG C, XIA L H, et al. Heterogeneous graph contrastive learning for recommendation[C]//Proceedings of the 16th ACM International Conference on Web Search and Data Mining. New York: ACM, 2023: 544-552.
[12] WU J C, WANG X, FENG F L, et al. Self-supervised graph learning for recommendation[C]//Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval. New York: ACM, 2021: 726-735.
[13] YU J, YIN H, XIA X, et al. Are graph augmentations necessary? Simple graph contrastive learning for recommendation[C]//Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval. New York: ACM, 2022: 1294-1303.
[14] XIA L H, HUANG C, XU Y, et al. Hypergraph contrastive collaborative filtering[C]//Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval. New York: ACM, 2022: 70-79.
[15] VAN DEN OORD A, LI Y Z, VINYALS O. Representation learning with contrastive predictive coding[J]. arXiv:1807. 03748, 2018.
[16] CHUANG C Y, HJELM R D, WANG X, et al. Robust contrastive learning against noisy views[C]//Proceedings of the 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2022: 16649-16660.
[17] WANG X, JI H Y, SHI C, et al. Heterogeneous graph attention network[C]//Proceedings of the World Wide Web Conference 2019. New York: ACM, 2019: 2022-2032.
[18] HU Z N, DONG Y X, WANG K S, et al. Heterogeneous graph transformer[C]//Proceedings of the Web Conference 2020. New York: ACM, 2020: 2704-2710.
[19] FU X Y, ZHANG J N, MENG Z Q, et al. MAGNN: metapath aggregated graph neural network for heterogeneous graph embedding[C]//Proceedings of the Web Conference 2020. New York: ACM, 2020: 2331-2341.
[20] YAN B, CAO Y, WANG H Y, et al. Federated heterogeneous graph neural network for privacy-preserving recommendation[C]//Proceedings of the ACM Web Conference 2024. New York: ACM, 2024: 3919-3929.
[21] JIANG Z, LIU H, FU B, et al. Recommendation in heterogeneous information networks based on generalized random walk model and Bayesian personalized ranking[C]//Proceedings of the 11th ACM International Conference on Web Search and Data Mining. New York: ACM, 2018: 288-296.
[22] SHI C, ZHANG Z Q, LUO P, et al. Semantic path based personalized recommendation on weighted heterogeneous information networks[C]//Proceedings of the 24th ACM International Conference on Information and Knowledge Management. New York: ACM, 2015: 453-462.
[23] SHI C, HU B B, ZHAO W X, et al. Heterogeneous information network embedding for recommendation[J]. IEEE Transactions on Knowledge and Data Engineering, 2019, 31(2): 357-370.
[24] GROVER A, LESKOVEC J. node2vec: scalable feature learning for networks[C]//Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM, 2016: 855-864.
[25] YU X, REN X, SUN Y Z, et al. Personalized entity recommendation: a heterogeneous information network approach[C]//Proceedings of the 7th ACM International Conference on Web Search and Data Mining. New York: ACM, 2014: 283-292.
[26] XIA L H, XU Y, HUANG C, et al. Graph meta network for multi-behavior recommendation[C]//Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval. New York: ACM, 2021: 757-766.
[27] XUAN S Y, ZHANG S L. Decoupled contrastive learning for long-tailed recognition[J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2024, 38(6): 6396-6403.
[28] CHEN T, KORNBLITH S, NOROUZI M, et al. A simple framework for contrastive learning of visual representations[C]//Proceedings of the 37th International Conference on Machine Learning, 2020: 1597-1607.
[29] LI J N, ZHOU P, XIONG C M, et al. Prototypical contrastive learning of unsupervised representations[J]. arXiv:2005. 04966, 2020.
[30] JING L, VINCENT P, LECUN Y, et al. Understanding dimensional collapse in contrastive self-supervised learning[J]. arXiv:2110.09348, 2021.
[31] HE X N, DENG K, WANG X, et al. LightGCN: simplifying and powering graph convolution network for recommendation[C]//Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval. New York: ACM, 2020: 639-648.
[32] PAN F Y, LI S K, AO X, et al. Warm up cold-start advertisements: improving CTR predictions via learning to learn ID embeddings[C]//Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval. New York: ACM, 2019: 695-704.
[33] RAJWADE A, RANGARAJAN A, BANERJEE A. Image denoising using the higher order singular value decomposition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(4): 849-862.
[34] HALKO N, MARTINSSON P G, TROPP J A. Finding structure with randomness: probabilistic algorithms for constructing approximate matrix decompositions[J]. SIAM Review, 2011, 53(2): 217-288.
[35] CAI X H, HUANG C, XIA L H, et al. LightGCL: simple yet effective graph contrastive learning for recommendation[J]. arXiv:2302.08191, 2023.
[36] RENDLE S, FREUDENTHALER C, GANTNER Z, et al. BPR: Bayesian personalized ranking from implicit feedback[J]. arXiv:1205.2618, 2012.
[37] HU B B, SHI C, ZHAO W X, et al. Leveraging meta-path based context for top-N recommendation with a neural co-attention model[C]//Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM, 2018: 1531-1540.
[38] WANG X, LIU N, HAN H, et al. Self-supervised heterogeneous graph neural network with co-contrastive learning[C]//Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. New York: ACM, 2021: 1726-1736.
[39] YANG L W, WANG S J, TAO Y Z, et al. DGRec: graph neural network for recommendation with diversified embedding generation[C]//Proceedings of the 16th ACM International Conference on Web Search and Data Mining. New York: ACM, 2023: 661-669.
[40] LONG X L, HUANG C, XU Y, et al. Social recommendation with self-supervised metagraph informax network[C]//Proceedings of the 30th ACM International Conference on Information and Knowledge Management. New York: ACM, 2021: 1160-1169.