Abstract
One of the biggest challenges in Recommendation Algorithms (RA) is how to obtain user and item embeddings from sparse interaction history. To take this challenge, most graph neural network based RAs explicitly incorporate high-order collaborative filtering signals on the user-item bipartite graph with either multi-layer semantics on the Knowledge Graph (KG) or multi-level neighbors on the social network. However, none of them fully integrate these three types of graph-structured data, which decreases embeddings’ precision. Based on this consideration, this paper integrates the three types of data by proposing a knowledge-rich influence propagation RA based on the graph attention mechanism. Specifically, in the semantic propagation, we categorize user preferences into deep interest obtained by multiple graph attention message propagations on related KG parts, and shallow interest generated from the interaction history. Moreover, the influence weight between items is determined by the number of co-interactions and the semantic similarity. These two factors as well as social relations together decide the influence weight between users. With these influence weights, final user and item embeddings are calculated through multi-layer message propagation. The experimental results show that the proposed recommendation algorithm outperforms several compelling baselines on six scaled-down real-world datasets. This work has confirmed the effectiveness of combining these three types of data to increase RAs’ coverage and accuracy.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Chen, T., Guo, J., et al.: Graph representation learning for popularity prediction problem: a survey. Discrete Math., Algorithms Appl. 14(7), 2230003 (2022)
Fan, W., Ma, Y., et al.: A graph neural network framework for social recommendations. TKDE 34(5), 2033–2047 (2020)
Gao, C., Wang, X., et al.: Graph neural networks for recommender system. In: WSDM, pp. 1623–1625 (2022)
Guo, Q., Zhuang, F., et al.: A survey on knowledge graph-based recommender systems. TKDE 34(8), 3549–3568 (2020)
He, X., Deng, K., et al.: LightGCN: simplifying and powering graph convolution network for recommendation. In: SIGIR, pp. 639–648 (2020)
Huang, C.: Recent advances in heterogeneous relation learning for recommendation. In: IJCAI, pp. 4442–4449 (2021)
Jiang, W., Sun, Y.: Social-RippleNet: jointly modeling of ripple net and social information for recommendation. Appl. Intell. 53(3), 3472–3487 (2022). https://doi.org/10.1007/s10489-022-03620-2
Li, A., Yang, B.: GSIRec: Learning with graph side information for recommendation. World Wide Web 24(5), 1411–1437 (2021). https://doi.org/10.1007/s11280-021-00910-6
Li, N., Gao, C., et al.: Disentangled modeling of social homophily and influence for social recommendation. TKDE 35(6), 5738–5751 (2022)
Liu, T., Wang, Z., et al.: Recommender systems with heterogeneous side information. In: WWW, pp. 3027–3033 (2019)
Liu, X., Yan, M., et al.: Sampling methods for efficient training of graph convolutional networks: a survey. IEEE/CAA J. Automatica Sinica 9(2), 205–234 (2022)
Rendle, S., Freudenthaler, C., et al.: BPR: bayesian personalized ranking from implicit feedback. In: UAI, pp. 452–461 (2009)
Wang, H., Zhang, F., et al.: RippleNet: propagating user preferences on the knowledge graph for recommender systems. In: CIKM, pp. 417–426 (2018)
Wang, H., Zhang, F., et al.: Multi-task feature learning for knowledge graph enhanced recommendation. In: WWW, pp. 2000–2010 (2019)
Wang, H., Zhao, M., et al.: Knowledge graph convolutional networks for recommender systems. In: WWW, pp. 3307–3313 (2019)
Wang, X., He, X., et al.: KGAT: knowledge graph attention network for recommendation. In: SIGKDD, pp. 950–958 (2019)
Wang, X., Huang, T., et al.: Learning intents behind interactions with knowledge graph for recommendation. In: WWW, pp. 878–887 (2021)
Wang, X., Wang, D., et al.: Explainable reasoning over knowledge graphs for recommendation. In: AAAI, pp. 5329–5336 (2019)
Wang, Z., Lin, G., et al.: CKAN: collaborative knowledge-aware attentive network for recommender systems. In: SIGIR, pp. 219–228 (2020)
Wang, Z., Wang, Z., et al.: Exploring multi-dimension user-item interactions with attentional knowledge graph neural networks for recommendation. IEEE Trans. Big Data, TBD 9(1), 212–226 (2023)
Wu, L., Li, J., et al.: DiffNet++: a neural influence and interest diffusion network for social recommendation. TKDE 34(10), 4753–4766 (2020)
Wu, L., Sun, P., et al.: A neural influence diffusion model for social recommendation. In: SIGIR, pp. 235–244 (2019)
Xu, X., Feng, W., et al.: Sampling methods for efficient training of graph convolutional networks: a survey. In: ICLR (2020)
Yu, J., Yin, H., et al.: Socially-aware self-supervised tri-training for recommendation. In: SIGKDD, pp. 2084–2092 (2021)
Zhang, F., Yuan, N.J., et al.: Collaborative knowledge base embedding for recommender systems. In: SIGKDD, pp. 353–362 (2016)
Zhang, S., Yao, L., et al.: Deep learning based recommender system: a survey and new perspectives. ACM Comput. Surv. 52(1), 1–38 (2019)
Zhu, F., Wang, Y., et al.: Cross-domain recommendation: challenges, progress, and prospects. In: IJCAI, pp. 4721–4728 (2021)
Acknowledgments
We are grateful to the reviewers of this paper for their constructive and insightful comments. The research reported in this paper was partially supported by the National Key R&D Program of China (NO. 2021YFB0300104), and the ANR project AGAPE ANR-18-CE23-0013.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Yang, Y., Jiang, G., Zhang, Y. (2023). Knowledge-Rich Influence Propagation Recommendation Algorithm Based on Graph Attention Networks. In: Yang, X., et al. Advanced Data Mining and Applications. ADMA 2023. Lecture Notes in Computer Science(), vol 14176. Springer, Cham. https://doi.org/10.1007/978-3-031-46661-8_10
Download citation
DOI: https://doi.org/10.1007/978-3-031-46661-8_10
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-46660-1
Online ISBN: 978-3-031-46661-8
eBook Packages: Computer ScienceComputer Science (R0)