Multi-Perspective Graph Contrastive Learning Recommendation Method with Layer Attention Mechanism

Qian Zhongsheng; Huang Heng; Zhu Hui; Liu Jinping

doi:10.7544/issn1000-1239.202330804

Journal of Computer Research and Development > 2025 > 62(1): 160-178. > DOI: 10.7544/issn1000-1239.202330804 CSTR: 32373.14.issn1000-1239.202330804

Qian Zhongsheng, Huang Heng, Zhu Hui, Liu Jinping. Multi-Perspective Graph Contrastive Learning Recommendation Method with Layer Attention Mechanism[J]. Journal of Computer Research and Development, 2025, 62(1): 160-178. DOI: 10.7544/issn1000-1239.202330804

Citation:

PDF (2217 KB)

Multi-Perspective Graph Contrastive Learning Recommendation Method with Layer Attention Mechanism

School of Computer and Artificial Intelligence, Jiangxi University of Finance and Economics, Nanchang 330013

Funds: This work was supported by the National Natural Science Foundation of China (62262025), the Key Program of the Natural Science Foundation of Jiangxi Province of China (20224ACB202012), and Jiangxi Poyang Support Program for Talents of China-Major Discipline Academic and Technical Leaders Training Program of Jiangxi Province-Leading Talent (Academic) (20243BCE51024).

More Information

Author Bio:
Qian Zhongsheng: born in 1977. PhD, professor, PhD supervisor. Senior member of CCF. His main research interests include machine learning, software engineering, and intelligent software

Huang Heng: born in 1999. Master. His main research interests include intelligent recommender system and software engineering

Zhu Hui: born in 1976. PhD candidate. His main research interests include intelligent recommender system and software engineering

Liu Jinping: born in 1995. PhD candidate. His main research interests include software engineering and intelligent recommender system
Received Date: October 09, 2023
Revised Date: April 22, 2024
Accepted Date: May 29, 2024
Available Online: June 30, 2024

Graphical Abstract

Abstract

Abstract

Graph contrastive learning is widely employed in recommender system due to its effectiveness in mitigating data sparsity issue. However, most current recommendation algorithms based on graph contrastive learning start to learn from only a single perspective, severely limiting the model’s generalization capability. Furthermore, the over-smoothing problem inherent in graph convolutional networks also affects the model’s stability. Based on this, we propose the multi-perspective graph contrastive learning recommendation method with layer attention mechanism. On the one hand, this method proposes three contrastive learning approaches from two different perspectives. From a view-level perspective, it constructs perturbation-enhanced view by adding random noise for the original graph and employing singular value decomposition (SVD) recombination to establish SVD-enhanced view. It then performs view-level contrastive learning on these two enhanced views. From a node-level perspective, it conducts contrastive learning on candidate nodes and candidate structural neighbors using semantic information between nodes, optimizes multi-task learning with three contrastive auxiliary tasks and a recommendation task to enhance the quality of node embeddings, thereby improving the model’s generalization ability. On the other hand, in the context of learning for user and item node embeddings by graph convolutional network, a layer attention mechanism is employed to aggregate the final node embeddings. This enhances the model’s higher-order connectivity and mitigates the over-smoothing issue. When compared with ten classic models on four publicly available datasets, such as LastFM, Gowalla, Ifashion, and Yelp, the results indicate that this method achieves an average improvement of 3.12% in Recall, 3.22% in Precision, and 4.06% in NDCG. This demonstrates the effectiveness of the approach proposed in this work.
- layer attention mechanism,
- contrastive learning,
- graph convolutional network,
- multi-task learning,
- recommender system

FullText(HTML)

References (34)

References

[1]	Chen Huiyuan, Lai V, Jin Hongye, et al. Towards mitigating dimensional collapse of representations in collaborative filtering[C]//Proc of the 17th ACM Int Conf on Web Search and Data Mining. New York: ACM, 2024: 106−115
[2]	Herlocker J L, Konstan J A, Terveen L G, et al. Evaluating collaborative filtering recommender systems[J]. ACM Transactions on Information Systems, 2004, 22(1): 5−53 doi: 10.1145/963770.963772
[3]	He Xiangnan, Liao Lizi, Zhang Hanwang, et al. Neural collaborative filtering[C]//Proc of the 26th Int Conf on World Wide Web. New York: ACM, 2017: 173−182
[4]	Rendle S, Freudenthaler C, Gantner Z, et al. BPR: Bayesian personalized ranking from implicit feedback[C]//Proc of the 25th Conf on Uncertainty in Artificial Intelligence. New York: ACM, 2009: 452−461
[5]	Wang Xiang, He Xiangnan, Wang Meng, et al. Neural graph collaborative filtering[C]//Proc of the 42nd Int ACM SIGIR Conf on Research and Development in Information Retrieval. New York: ACM, 2019: 165−174
[6]	He Xiangnan, Deng Kuan, Wang Xiang, et al. LightGCN: Simplifying and powering graph convolution network for recommendation[C]//Proc of the 43rd Int ACM SIGIR Conf on Research and Development in Information Retrieval. New York: ACM, 2020: 639−648
[7]	钱忠胜,赵畅,俞情媛,等. 结合注意力CNN与GNN的信息融合推荐方法[J]. 软件学报,2023,34(5):2317−2336 Qian Zhongsheng, Zhao Chang, Yu Qingyuan, et al. Information fusion recommendation approach combining attention CNN and GNN[J]. Journal of Software, 2023, 34(5): 2317−2336 (in Chinese)
[8]	任豪,刘柏嵩,孙金杨,等. 基于时间和关系感知的图协同过滤跨域序列推荐[J]. 计算机研究与发展,2023,60(1):112−124 Ren Hao, Liu Baisong, Sun Jinyang, et al. A time and relation-aware graph collaborative filtering for cross-domain sequential recommendation[J]. Journal of Computer Research and Development, 2023, 60(1): 112−124(in Chinese)
[9]	Wu Jiancan, Wang Xiang, Feng Fuli, et al. Self-supervised graph learning for recommendation[C]//Proc of the 44th Int ACM SIGIR Conf on Research and Development in Information Retrieval. New York: ACM, 2021: 726−735
[10]	He Wei, Sun Guohao, Lu Jinhu, et al. Candidate-aware graph contrastive learning for recommendation[C]//Proc of the 46th Int ACM SIGIR Conf on Research and Development in Information Retrieval. New York: ACM, 2023: 1670−1679
[11]	Yu Junliang, Yin Hongzhi, Xia Xin, et al. Are graph augmentations necessary? Simple graph contrastive learning for recommendation[C]//Proc of the 45th Int ACM SIGIR Conf on Research and Development in Information Retrieval. New York: ACM, 2022: 1294−1303
[12]	Xia Lianghao, Huang Chao, Xu Yong, et al. Hypergraph contrastive collaborative filtering[C]//Proc of the 45th Int ACM SIGIR Conf on Research and Development in Information Retrieval. New York: ACM, 2022: 70−79
[13]	Lin Zihan, Tian Changxin, Hou Yupeng, et al. Improving graph collaborative filtering with neighborhood-enriched contrastive learning[C]//Proc of the 31st Int Conf on World Wide Web. New York: ACM, 2022: 2320−2329
[14]	Liu Fan, Cheng Zhiyong, Zhu Lei, et al. Interest-aware message-passing GCN for recommendation[C]//Proc of the 30th Int Conf on World Wide We. New York: ACM, 2021: 1296−1305
[15]	Gao Chen, Wang Xiang, He Xiangnan, et al. Graph neural networks for recommender system[C]//Proc of the 15th ACM Int Conf on Web Search and Data Mining. New York: ACM, 2022: 1623−1625
[16]	Wu Zonghan, Pan Shirui, Chen Fengwen, et al. A comprehensive survey on graph neural networks[J]. IEEE Transactions on Neural Networks and Learning Systems, 2021, 32(1): 4−24 doi: 10.1109/TNNLS.2020.2978386
[17]	Kipf T N, Welling M. Semi-supervised classification with graph convolutional networks[C/OL]//Proc of the 5th Int Conf on Learning Representations. 2017[2024-02-25]. https://openreview.net/forum?id=SJU4ayYgl
[18]	闫昭,项欣光,李泽超. 基于交互序列商品相关性建模的图卷积会话推荐[J]. 中国科学:信息科学,2022,52(6):1069−1082 doi: 10.1360/SSI-2020-0383 Yan Zhao, Xiang Xinguang, Li Zechao. Item correlation modeling in interaction sequence for graph convolutional session recommendation[J]. SCIENTIA SINICA Informationis, 2022, 52(6): 1069−1082 (in Chinese) doi: 10.1360/SSI-2020-0383
[19]	李挺,金福生,李荣华,等. Light-HGNN:用于圈层内容推荐的轻量同质超图神经网络[J]. 计算机研究与发展,2024,61(4):877−888 Li Ting, Jin Fusheng, Li Ronghua, et al. Light-HGNN: Lightweight homogeneous hypergraph neural network for circle content recommendation[J]. Journal of Computer Research and Development, 2024, 61(4): 877−888 (in Chinese)
[20]	Berg R, Kipf T N, Welling M. Graph convolutional matrix completion[C]//Proc of the 24th ACM SIGKDD Int Conf on Knowledge Discovery and Data Mining. New York: ACM, 2018: 974−983
[21]	Ji Shuyi, Feng Yifan, Ji Rongrong, et al. Dual channel hypergraph collaborative filtering[C]//Proc of the 26th ACM SIGKDD Int Conf on Knowledge Discovery & Data Mining. New York: ACM, 2020: 2020−2029
[22]	Chen Huiyuan, Yeh C M, Wang Fei, et al. Graph neural transport networks with non-local attentions for recommender systems[C]//Proc of the 31st Int Conf on World Wide Web. New York: ACM, 2022: 1955−1964
[23]	Huang Tinglin, Dong Yuxiao, Ding Ming, et al. MixGCF: An improved training method for graph neural network-based recommender systems[C]//Proc of the 27th ACM SIGKDD Int Conf on Knowledge Discovery & Data Mining. New York: ACM, 2021: 665−674
[24]	Mao Kelong, Zhu Jieming, Xiao Xi, et al. UltraGCN: Ultra simplification of graph convolutional networks for recommendation[C]//Proc of the 30th ACM Int Conf on Information & Knowledge Management. New York: ACM, 2021: 1253−1262
[25]	Shen Yifei, Wu Yongji, Zhang Yao, et al. How powerful is graph convolution for recommendation?[C]//Proc of the 30th ACM Int Conf on Information & Knowledge Management. New York: ACM, 2021: 1619−1629
[26]	Zhang Xiaoyu, Xin Xin, Li Dongdong, et al. Variational reasoning over incomplete knowledge graphs for conversational recommendation[C]//Proc of the 16th ACM Int Conf on Web Search and Data Mining. New York: ACM, 2023: 231−239
[27]	曹阳,高旻,余俊良,等. 基于双图混合随机游走的社会化推荐模型[J]. 电子学报,2023,51(2):286−296 Cao Yang, Gao Min, Yu Junliang, et al. Bi-graph mix-random walk based social recommendation model[J]. Acta Electronica Sinica, 2023, 51(2): 286−296 (in Chinese)
[28]	Yan Mingshi, Cheng Zhiyong, Gao Chen, et al. Cascading residual graph convolutional network for multi-behavior recommendation[J]. ACM Transactions on Information Systems, 2023, 42(1): 1−26
[29]	He Kaiming, Fan Haoqi, Wu Yuxin, et al. Momentum contrast for unsupervised visual representation learning[C]//Proc of the 33rd IEEE/CVF Conf on Computer Vision and Pattern Recognition. Piscataway, NJ: IEEE, 2020: 9729−9738
[30]	Cai Xuheng, Huang Chao, Xia Lianghao, et al. LightGCL: Simple yet effective graph contrastive learning for recommendation[C/OL]//Proc of the 11th Int Conf of Learning Representation. 2023[2024-02-25]. https://openreview.net/forum?id=FKXVK9dyMM
[31]	Xia Lianghao, Huang Chao, Huang Chunzhen, et al. Automated self-supervised learning for recommendation[C]//Proc of the 32nd Int Conf on World Wide Web. New York: ACM, 2023: 992−1002
[32]	Li Chaoliu, Xia Lianghao, Ren Xubin, et al. Graph transformer for recommendation[C]//Proc of the 46th Int ACM SIGIR Conf on Research and Development in Information Retrieval. New York: ACM, 2023: 1680−1689
[33]	Wang Chenyang, Yu Yuanqing, Ma Weizhi, et al. Towards representation alignment and uniformity in collaborative filtering[C]//Proc of the 28th ACM SIGKDD Conf on Knowledge Discovery & Data Mining. New York: ACM, 2022: 1816−1825
[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 doi: 10.1137/090771806