Graph Contrastive Learning with Hybrid Noise Augmentation for Recommendation
Pages 325 - 339
Abstract
Recommendation System is one of the effective tools to solve the problem of information overload in the era of big data, but the data sparsity has greatly affected its performance. Recently, contrastive learning, has attracted great attention and is expected to solve this problem. However, most of the existing graph-based contrastive learning methods perturb the original graph for data enhancement, which may affect the recommended performance. Meanwhile, studies have shown that improving the uniformity of data distribution is more important than data augmentation by graph perturbation. In this paper, in order to improve the uniformity of the data distribution, we propose a Graph Contrastive Learning with Hybrid Noise Augmentation for Recommendation, which is abbreviated as GCLHANRec. Specifically, we add uniform distribution random noise to users and normal distribution random noise to items, to improve the data uniformity while increasing the user’s interest diversity for different items, thereby improving the accuracy and personalization degree of the recommendation system. Additionally, we propose Balanced Bayesian Personalized Ranking (BBPR) as the loss function for recommendation tasks, which is a modification of BPR to better make the model pay more attention to the difference between positive and negative samples, thus performing better in ranking tasks. We conducted extensive experiments using three datasets collected from actual environment, including Movielens, LastFM and Douban-book. The results show that our method outperforms several existing methods.
References
[1]
Li Y, Zhou T, Yang K, et al. Personalized recommender systems based on social relationships and historical behaviors Appl. Math. Comput. 2023 437
[2]
Liu T, Wu Q, Chang L, et al. A review of deep learning-based recommender system in e-learning environments Artif. Intell. Rev. 2022 55 8 5953-5980
[3]
Roy D, Dutta M, et al. Optimal hierarchical attention network-based sentiment analysis for movie recommendation Soc. Netw. Anal. Min. 2022 12 1 138
[4]
Zhang, H., Wang, H., Wang, G., et al.: A hyperbolic-to-hyperbolic user representation with multi-aspect for social recommendation. In: ACM International Conference on Information & Knowledge Management, pp. 4667–4671 (2022)
[5]
Qiu Z, Hu Y, and Wu X Graph neural news recommendation with user existing and potential interest modeling ACM Trans. Knowl. Discov. Data 2022 16 5 96:1-96:17
[6]
Tao, Y., Li, Y., Zhang, S., et al.: Revisiting graph based social recommendation: a distillation enhanced social graph network. In: WWW, pp. 2830–2838 (2022)
[7]
Wang, J., Chen, Y., Wang, Z., et al.: Popularity-enhanced news recommendation with multi-view interest representation. In: CIKM, pp. 1949–1958 (2021)
[8]
Shi C, Han X, Song L, et al. Deep collaborative filtering with multi-aspect information in heterogeneous networks IEEE Trans. Knowl. Data Eng. 2021 33 4 1413-1425
[9]
Wang, X., He, X., Wang, M., et al.: Neural graph collaborative filtering. In: SIGIR, pp. 165–174 (2019)
[10]
Dong, X., Yu, L., Wu, Z., et al.: A hybrid collaborative filtering model with deep structure for recommender systems. In: AAAI, pp. 1309–1315 (2017)
[11]
Pérez-Almaguer, Y., Yera, R., Alzahrani, A.A., et al.: Content-based group recommender systems: a general taxonomy and further improvements. Expert Syst. Appl. 115444 (2021)
[12]
Peng, Y.: A survey on modern recommendation system based on big data. CoRR, abs/2206.02631 (2022)
[13]
Joshi, A., Wong, C., de Oliveira, D.M., et al.: Imbalanced data sparsity as a source of unfair bias in collaborative filtering. In: RecSys, pp. 531–533 (2022)
[14]
Elahi E and Halim Z Graph attention-based collaborative filtering for user-specific recommender system using knowledge graph and deep neural networks Knowl. Inf. Syst. 2022 64 9 2457-2480
[15]
Chen, T., Kornblith, S., Norouzi, M., Hinton, G.E.: A simple framework for contrastive learning of visual representations. In: PMLR, pp. 1597–1607 (2020)
[16]
Zeng, J., Xie, P.: Contrastive self-supervised learning for graph classification. In: AAAI, pp. 10824–10832 (2021)
[17]
You, Y., Chen, T., Sui, Y., et al.: Graph contrastive learning with augmentations. In: NeurIPS (2020)
[18]
Jaiswal, A., Babu, A.B., Zadeh, M.Z., et al.: A survey on contrastive self-supervised learning. CoRR, abs/2011.00362 (2020)
[19]
Hassani, K., Khas Ahmadi, A.H.: Contrastive multi-view representation learning on graphs. In: ICML, pp. 4116–4126 (2020)
[20]
Yu, J., Yin, H., Xia, X., et al.: Self-supervised learning for recommender systems: a survey. CoRR, abs/2203.15876 (2022)
[21]
Liu, Z., Ma, Y., Ouyang, Y., et al.: Contrastive learning for recommender system. CoRR, abs/2101.01317 (2021)
[22]
Yu, J., Yin, H., Gao, M., et al.: Socially-aware self-supervised tri-training for recommendation. In: KDD, pp. 2084–2092 (2021)
[23]
Wu, J., Wang, X., Feng, F., et al.: Self-supervised graph learning for recommendation. In: SIGIR, pp. 726–735 (2021)
[24]
Lin, Z., Tian, C., Hou, Y., et al.: Improving graph collaborative filtering with neighborhood-enriched contrastive learning. In: WWW, pp. 2320–2329 (2022)
[25]
Yu, J., Xia, X., Chen, T., et al.: XSimGCL: towards extremely simple graph contrastive learning for recommendation. CoRR, abs/2209.02544 (2022)
[26]
Zhang, J., Gao, M., Yu, J., et al.: Double-scale self-supervised hypergraph learning for group recommendation. In: CIKM, pp. 2557–2567 (2021)
[27]
Zhou, X., Sun, A., Liu, Y., et al.: SelfCF: a simple framework for self-supervised collaborative filtering. CoRR, abs/2107.03019 (2021)
[28]
Ying, R., He, R., Chen, K., et al.: Graph convolutional neural networks for web-scale recommender systems. In: KDD, pp. 974–983 (2018)
[29]
Xue F, He X, Wang X, et al. Deep item-based collaborative filtering for top-n recommendation ACM Trans. Inf. Syst. 2019 37 3 33:1-33:25
[30]
He, X., Deng, K., Wang, X., et al.: LightGCN: simplifying and powering graph convolution network for recommendation. In: SIGIR, pp. 639–648 (2020)
[31]
Chen, L., Wu, L., Hong, R., et al.: Revisiting graph based collaborative filtering: a linear residual graph convolutional network approach. In: IAAI, pp. 27–34 (2020)
[32]
Huang, T., Dong, Y., Ding, M., et al.: MixgCF: an improved training method for graph neural network-based recommender systems. In: KDD, pp. 665–674 (2021)
[33]
Yu, J., Yin, H., Xia, X., et al.: Graph augmentation-free contrastive learning for recommendation. CoRR, abs/2112.08679 (2021)
[34]
Rendle, S., Freudenthaler, C., Gantner, Z., et al.: BPR: Bayesian personalized ranking from implicit feedback. In: UAI, pp. 452–461 (2009)
[35]
Oord, A.V.D., Li, Y., Vinyals, O.: Representation learning with contrastive predictive coding. arXiv preprint arXiv:1807.03748 (2018)
[36]
Wang, F., Liu, H.: Understanding the behaviour of contrastive loss. In: CVPR, pp. 2495–2504 (2021)
[37]
Koren, Y., Bell, R.M., Volinsky, C.: Matrix factorization techniques for recommender systems. Computer 30–37 (2009)
[38]
Kingma, D.P., Ba, J.: Adam: a method for stochastic optimization. In: ICLR (2015)
Recommendations
A Review-aware Graph Contrastive Learning Framework for Recommendation
SIGIR '22: Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information RetrievalMost modern recommender systems predict users' preferences with two components: user and item embedding learning, followed by the user-item interaction modeling. By utilizing the auxiliary review information accompanied with user ratings, many of the ...
SGCCL: Siamese Graph Contrastive Consensus Learning for Personalized Recommendation
WSDM '23: Proceedings of the Sixteenth ACM International Conference on Web Search and Data MiningContrastive-learning-based neural networks have recently been introduced to recommender systems, due to their unique advantage of injecting collaborative signals to model deep representations, and the self-supervision nature in the learning process. ...
A novel learning-to-rank based hybrid method for book recommendation
WI '17: Proceedings of the International Conference on Web IntelligenceRecommendation system is able to recommend items that are likely to be preferred by the user. Hybrid recommender systems combine the advantages of the collaborative filtering and content-based filtering for improved recommendation. Hybrid recommendation ...
Comments
Please enable JavaScript to view thecomments powered by Disqus.Information & Contributors
Information
Published In
Aug 2023
716 pages
ISBN:978-3-031-46673-1
DOI:10.1007/978-3-031-46674-8
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023.
Publisher
Springer-Verlag
Berlin, Heidelberg
Publication History
Published: 05 November 2023
Author Tags
Qualifiers
- Article
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 0Total Downloads
- Downloads (Last 12 months)0
- Downloads (Last 6 weeks)0
Reflects downloads up to 01 Oct 2024
Other Metrics
Citations
View Options
View options
Get Access
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Sign in