Knowledge-aware recommendation with relation-denoising counterfactual generators

Abstract

Knowledge graph (KG), enriched with abundant auxiliary information, plays an increasingly important role in enhancing recommendation performance while simultaneously improving the system’s explainability. Recently, the development of end-to-end models leveraging graph neural networks (GNNs) has emerged as a leading trend in knowledge-aware recommendation systems. However, the noise data in dataset significantly impact the performance of recommendation systems, often leading to misleading associations between knowledge facts and user behaviors. In this paper, we propose a novel approach that leverages both causal inference and denoising techniques to address the challenge. Specifically, our approach uses two counterfactual generators, built with reinforcement learning, to create counterfactual interactions. The recommender is implemented using two distinct graph neural networks to aggregate information from the Knowledge Graph and user-item interactions, respectively. Additionally, we combine relation-denoising module that filtering out irrelevant relations for both the original and generated interaction graphs. With the help of counterfactual generators and denoising module, the recommender could identify potential spurious correlations and reduce the influence of noise. We conducted extensive experiments on three benchmark datasets, and the results demonstrate the effectiveness of our approach compared to state-of-the-art knowledge-aware methods.

Graphical abstract

Keywords

recommender system / knowledge graph / graph neural network

References

[1] Wang H, Zhang F, Wang J, Zhao M, Li W, Xie X, Guo M. RippleNet: propagating user preferences on the knowledge graph for recommender systems. In: Proceedings of the 27th ACM International Conference on Information and Knowledge Management. 2018, 417−426

[2] Zhang F, Yuan N J, Lian D, Xie X, Ma W Y. Collaborative knowledge base embedding for recommender systems. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2016, 353−362

[3] Wang X, He X, Cao Y, Liu M, Chua T S. KGAT: knowledge graph attention network for recommendation. In: Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. 2019, 950−958

[4] Zhu X, Du Y, Mao Y, Chen L, Hu Y, Gao Y. Knowledge-refined denoising network for robust recommendation. In: Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval. 2023, 362−371

[5] Huang S, Hu C, Kong W, Liu Y. Disentangled contrastive learning for knowledge-aware recommender system. In: Proceedings of the 22nd International Semantic Web Conference. 2023, 140−158

[6] Wang Z, Zhang J, Feng J, Chen Z. Knowledge graph embedding by translating on hyperplanes. In: Proceedings of the 28th AAAI Conference on Artificial Intelligence. 2014, 1112−1119

[7] Wang H, Zhang F, Xie X, Guo M. DKN: deep knowledge-aware network for news recommendation. In: Proceedings of 2018 World Wide Web Conference. 2018, 1835−1844

[8] Zhang J, Cai S, Jiang Z, Xiao J, Ming Z. FireRobBrain: planning for a firefighting robot using knowledge graph and large language model. In: Proceedings of the 10th IEEE International Conference on Intelligent Data and Security. 2024, 37−41

[9] Xia Y, Fang H, Zhang J, Long C. Leveraging knowledge graph embedding for effective conversational recommendation, 2024, arXiv preprint arXiv: 2408.01342

[10] Shokrzadeh Z, Feizi-Derakhshi M R, Balafar M A, Bagherzadeh Mohasefi J . Knowledge graph-based recommendation system enhanced by neural collaborative filtering and knowledge graph embedding. Ain Shams Engineering Journal, 2024, 15( 1): 102263

[11] Chen C, Zhang M, Ma W, Liu Y, Ma S. Jointly non-sampling learning for knowledge graph enhanced recommendation. In: Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval. 2020, 189−198

[12] Hu B, Shi C, Zhao W X, Yu P S. Leveraging meta-path based context for top- N recommendation with a neural co-attention model. In: Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. 2018, 1531−1540

[13] Huang J, Zhao W X, Dou H, Wen J R, Chang E Y. Improving sequential recommendation with knowledge-enhanced memory networks. In: Proceedings of the 41st International ACM SIGIR Conference on Research & Development in Information Retrieval. 2018, 505−514

[14] Jiang Y, Yang Y, Xia L, Huang C. DiffKG: knowledge graph diffusion model for recommendation. In: Proceedings of the 17th ACM International Conference on Web Search and Data Mining. 2024, 313−321

[15] Wang J, Xie H, Zhang S, Qin S J, Tao X, Wang F L, Xu X . Multimodal fusion framework based on knowledge graph for personalized recommendation. Expert Systems with Applications, 2025, 268: 126308

[16] Kipf T N, Welling M. Semi-supervised classification with graph convolutional networks. In: Proceedings of the 5th International Conference on Learning Representations. 2017

[17] Veličković P, Cucurull G, Casanova A, Romero A, Liò P, Bengio Y. Graph attention networks. 2017, arXiv preprint arXiv: 1710.10903

[18] Wu Z. An efficient recommendation model based on knowledge graph attention-assisted network (KGATAX). 2024, arXiv preprint arXiv: 2409.15315

[19] Sha X, Sun Z, Zhang J . Hierarchical attentive knowledge graph embedding for personalized recommendation. Electronic Commerce Research and Applications, 2021, 48: 101071

[20] Wang H, Zhang F, Zhang M, Leskovec J, Zhao M, Li W, Wang Z. Knowledge-aware graph neural networks with label smoothness regularization for recommender systems. In: Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. 2019, 968−977

[21] Wang X, Huang T, Wang D, Yuan Y, Liu Z, He X, Chua T S. Learning intents behind interactions with knowledge graph for recommendation. In: Proceedings of the Web Conference 2021. 2021, 878−887

[22] Alatrash R, Chatti M A, Ul Ain Q, Fang Y, Joarder S, Siepmann C . ConceptGCN: knowledge concept recommendation in MOOCs based on knowledge graph convolutional networks and SBERT. Computers and Education: Artificial Intelligence, 2023, 6: 100193

[23] Gao Y, Du Y, Hu Y, Chen L, Zhu X, Fang Z, Zheng B. Self-guided learning to denoise for robust recommendation. In: Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval. 2022, 1412−1422

[24] Pearl J. Causality. 2nd ed. Cambridge: Cambridge University Press, 2009

[25] Mu S, Li Y, Zhao W X, Wang J, Ding B, Wen J R. Alleviating spurious correlations in knowledge-aware recommendations through counterfactual generator. In: Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval. 2022, 1401−1411

[26] Roese N J . Counterfactual thinking. Psychological Bulletin, 1997, 121( 1): 133–148

[27] Wang Z, Zhang J, Xu H, Chen X, Zhang Y, Zhao W X, Wen J R. Counterfactual data-augmented sequential recommendation. In: Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval. 2021, 347−356

[28] Yang M, Dai Q, Dong Z, Chen X, He X, Wang J. Top-N recommendation with counterfactual user preference simulation. In: Proceedings of the 30th ACM International Conference on Information & Knowledge Management. 2021, 2342−2351

[29] Wang W, Feng F, He X, Zhang H, Chua T S. Clicks can be cheating: counterfactual recommendation for mitigating clickbait issue. In: Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval. 2021, 1288−1297

[30] Ai Q, Azizi V, Chen X, Zhang Y . Learning heterogeneous knowledge base embeddings for explainable recommendation. Algorithms, 2018, 11( 9): 137

[31] Mu S, Li Y, Zhao W X, Li S, Wen J R . Knowledge-guided disentangled representation learning for recommender systems. ACM Transactions on Information Systems (TOIS), 2021, 40( 1): 6

[32] Cao Y, Wang X, He X, Hu Z, Chua T S. Unifying knowledge graph learning and recommendation: Towards a better understanding of user preferences. In: Proceedings of the World Wide Web Conference. 2019, 151−161

[33] Wang H, Zhang F, Zhao M, Li W, Xie X, Guo M. Multi-task feature learning for knowledge graph enhanced recommendation. In: Proceedings of the World Wide Web Conference. 2019, 2000−2010

[34] Wang H, Zhao M, Xie X, Li W, Guo M. Knowledge graph convolutional networks for recommender systems. In: Proceedings of the World Wide Web Conference. 2019, 3307−3313

[35] Li Y, Wang D, Sun Z, Zhang H, Guo H. LightKG: efficient knowledge-aware recommendations with simplified GNN architecture. In: Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.2. 2025, 1577−1588

[36] Gupta P, Sharma A, Malhotra P, Vig L, Shroff G. CauSeR: causal session-based recommendations for handling popularity bias. In: Proceedings of the 30th ACM International Conference on Information & Knowledge Management. 2021, 3048−3052

[37] Zhang S, Yao D, Zhao Z, Chua T S, Wu F. CauseRec: counterfactual user sequence synthesis for sequential recommendation. In: Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval. 2021, 367−377

[38] Wang Z, Xu Q, Yang Z, Cao X, Huang Q. Implicit feedbacks are not always favorable: iterative relabeled one-class collaborative filtering against noisy interactions. In: Proceedings of the 29th ACM International Conference on Multimedia. 2021, 3070−3078

[39] Wang Y, Xin X, Meng Z, Jose J M, Feng F, He X. Learning robust recommenders through cross-model agreement. In: Proceedings of the ACM Web Conference 2022. 2022, 2015−2025

[40] Chen H, Wang L, Lin Y, Yeh C C M, Wang F, Yang H. Structured graph convolutional networks with stochastic masks for recommender systems. In: Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval. 2021, 614−623

[41] Sutton R S, Barto A G. Reinforcement Learning: An Introduction. Cambridge: A Bradford Book, 2018

[42] Dong Z, Mnih A, Tucker G. DisARM: an antithetic gradient estimator for binary latent variables. In: Proceedings of the 34th International Conference on Neural Information Processing Systems. 2020, 1565

[43] Hamilton W L, Ying Z, Leskovec J. Inductive representation learning on large graphs. In: Proceedings of the 31st International Conference on Neural Information Processing Systems. 2017, 1025−1035

[44] Tishby N, Pereira F C, Bialek W. The information bottleneck method. 2000, arXiv preprint arXiv: physics/0004057

[45] Alemi A A, Fischer I, Dillon J V, Murphy K. Deep variational information bottleneck. In: Proceedings of the 5th International Conference on Learning Representations. 2017

[46] Rendle S, Freudenthaler C, Gantner Z, Schmidt-Thieme L. BPR: Bayesian personalized ranking from implicit feedback. 2012, arXiv preprint arXiv: 1205.2618

[47] Harper F M, Konstan J A . The MovieLens datasets: history and context. ACM Transactions on Interactive Intelligent Systems (TiIS), 2016, 5( 4): 19

[48] Schedl M. The LFM-1b dataset for music retrieval and recommendation. In: Proceedings of 2016 ACM on International Conference on Multimedia Retrieval. 2016, 103−110

[49] Zhao W X, He G, Yang K, Dou H, Huang J, Ouyang S, Wen J R . KB4Rec: a data set for linking knowledge bases with recommender systems. Data Intelligence, 2019, 1( 2): 121–136

[50] Zhao W X, Mu S, Hou Y, Lin Z, Chen Y, Pan X, Li K, Lu Y, Wang H, Tian C, Min Y, Feng Z, Fan X, Chen X, Wang P, Ji W, Li Y, Wang X, Wen J R. RecBole: towards a unified, comprehensive and efficient framework for recommendation algorithms. In: Proceedings of the 30th ACM International Conference on Information & Knowledge Management. 2021, 4653−4664

RIGHTS & PERMISSIONS

Higher Education Press