融合图神经网络和注意力机制的会话推荐模型
Session-Based Recommendation Model with Graph Neural Network and Attention Mechanism
DOI: 10.12677/CSA.2022.124114, PDF,    科研立项经费支持
作者: 黄浩文, 陈平华：广东工业大学，计算机学院，广东 广州
关键词: 会话推荐；图神经网络；门控循环单元；注意力机制；Session-Based Recommendation； Graph Neural Network； Gated Recurrent Unit； Attention Mechanism
摘要: 针对电子商务场景的会话推荐难以处理用户行为随机性、商品数据稀疏性和推荐结果滞后性问题，提出融合图神经网络和注意力机制的会话推荐模型。首先引入相对点击时间率改善由用户随机点击造成的推荐性能下降问题，参与生成由会话转换的商品关系依赖图结构；然后由多层门控图神经网络处理图结构，通过聚合更多节点信息输出相对稠密的商品表示；接着使用门控循环单元捕捉会话信息，并借助注意力机制强化会话靠后的项目，综合形成用户表示，最终获得实时推荐。模型在Yoochoose和Diginetica两个公开数据集上进行实验，获得了较好的表现，结果表明所提出的模型可以提高推荐准确性。
Abstract: Aiming at the problems that session-based recommendation in e-commerce platforms hardly solve the problems of random-like behavior from users, the sparsity from items and the real-time performance from recommendation results, a session-based recommendation model with graph neural network and attention mechanism is proposed. First, the relative click time rate is introduced to tackle the recommendation effect degradation caused by random clicks from users, and partici-pates in the generation of the item relationship dependency graph transformed by sessions. Then, the multi-layer gated graph neural network is used for learning the graph structure, and the relatively dense item’s representation is output by aggregating more node information. After that, each session’s information is captured by gated recurrent units, and items at the rear of the session are strengthened with the help of the attention mechanism to generate each user’s representation, so as to finally obtain real-time recommendation. Experiments on Yoochoose and Diginetica datasets show that the proposed model can achieve good performance and improve the accuracy of recommendation.
文章引用：黄浩文, 陈平华. 融合图神经网络和注意力机制的会话推荐模型[J]. 计算机科学与应用, 2022, 12(4): 1108-1121. https://doi.org/10.12677/CSA.2022.124114

参考文献

[1] Wang, S., Cao, L., Wang, Y., et al. (2022) A Survey on Session-Based Recommender Systems. ACM Computing Sur-veys, 54, Article No. 154. [Google Scholar] [CrossRef] 
[2] Nilashi, M., Bagherifard, K., Ibrahim, O., et al. (2013) Collaborative Filtering Recommender Systems. Research Journal of Applied Sciences, Engineering and Technol-ogy, 5, 4168-4182. [Google Scholar] [CrossRef] 
[3] Koren, Y., Bell, R. and Volinsky, C. (2009) Matrix Factorization Techniques for Recommender Systems. IEEE Computer Journal, 42, 30-37. [Google Scholar] [CrossRef] 
[4] Rendle, S., Freudenthaler, C. and Schmidt-Thieme, L. (2010) Factoriz-ing Personalized Markov Chains for Next-Basket Recommendation. Proceedings of the 19th International Conference on World Wide Web, Raleigh, 26-30 April 2010, 811-820. [Google Scholar] [CrossRef] 
[5] 黄立威, 江碧涛, 吕守业, 等. 基于深度学习的推荐系统研究综述[J]. 计算机学报, 2018, 41(7): 1619-1647.
[6] Scarselli, F., Gori, M., Tsoi, A.C., et al. (2009) The Graph Neural Network Model. IEEE Transactions on Neural Networks, 20, 61-80. [Google Scholar] [CrossRef] 
[7] Sarwar, B., Karypis, G., Konstan, J., et al. (2001) Item-Based Collaborative Filtering Recommendation. In: Proceedings of the 10th International Conference on World Wide Web, Association for Computing Machinery, New York, 285-295. [Google Scholar] [CrossRef] 
[8] Davidson, J., Liebald, B., Liu, J., et al. (2010) The YouTube Video Recommendation System. RecSys’10: Proceedings of the 4th ACM conference on Recommender Systems, Association for Computing Machinery, New York, 293-296. [Google Scholar] [CrossRef] 
[9] Bamshad, M., Honghua, D., Tao, L., et al. (2001) Effective Person-alization Based on Association Rule Discovery from Web Usage Data. Proceedings of the 3rd ACM Workshop on Web Information and Data Management, Atlanta, November 2001, 9-15.
[10] Rendle, S., Freudenthaler, C., Gantner, Z., et al. (2009) BPR: Bayesian Personalized Ranking from Implicit Feedback. In: UAI’09: Proceedings of the Twenty-Fifth Conference on Uncertainty in Artificial Intelligence, AUAI Press, Arlington, 452-461.
[11] Bahdanau, D., Cho, K. and Bengio, Y. (2014) Neural Machine Translation by Jointly Learning to Align and Translate. Computer Science.
[12] He, Y., Sainath, T.N., Prabhavalkar, R., et al. (2018) Streaming End-to-End Speech Recognition for Mobile Devices. 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Brighton, 12-17 May 2019, 6381-6385. [Google Scholar] [CrossRef] 
[13] Zhou, L., Zhou, Y., Corso, J., et al. (2018) End-to-End Dense Video Captioning with Masked Transformer. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Salt Lake City, 18-23 June 2018, 8739-8748. [Google Scholar] [CrossRef] 
[14] Balázs, H., Alexandros, K., et al. (2016) Session-Based Recom-mendations with Recurrent Neural Networks. Proceedings of the 4th International Conference on Learning Representa-tions (ICLR 2016), San Juan, 2-4 May 2016, 1-10. [Google Scholar] [CrossRef] 
[15] Li, J., Ren, P., Chen, Z., et al. (2017) Neural Attentive Ses-sion-Based Recommendation. In: Proceeding of the 11th ACM International Conference on Information and Knowledge Management, Association for Computing Machinery, New York, 1418. [Google Scholar] [CrossRef] 
[16] Vaswani, A., Shazeer, N., Parmar, N., et al. (2017) Attention Is All You Need. 31st Conference on Neural Information Processing Systems (NIPS 2017), Long Beach, 4-9 December 2017, 6000-6010.
[17] Liu, Q., Zeng, Y., Mokhosi, R., et al. (2018) STAMP: Short-Term Attention/Memory Priority Model for Session-Based Recommendation. In: KDD’18: Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, Association for Computing Machinery, New York, 1831-1839. [Google Scholar] [CrossRef] 
[18] Kipf, T.N. and Welling, M. (2017) Semi-Supervised Classification with Graph Convolutional Networks. 5th International Conference on Learning Representations, ICLR 2017—Conference Track Proceedings, Toulon, 24-26 April 2017, 1-14. [Google Scholar] [CrossRef] 
[19] Hamilton, W., Ying, Z. and Leskovec, J. (2017) Inductive Repre-sentation Learning on Large Graphs. In: 31st Conference on Neural Information Processing Systems (NIPS 2017), Cur-ran Associates, Inc., Long Beach, 1-11. [Google Scholar] [CrossRef] 
[20] Li, Y., Tarlow, D., Brockschmidt, M., et al. (2016) Gated Graph Sequence Neural Networks. Proceedings of the 4th International Conference on Learning Representations (ICLR 2016), Caribe Hilton, 2-4 May 2016, 1-20. [Google Scholar] [CrossRef] 
[21] Yehuda, K., Robert, B. and Chris, V. (2009) Matrix Factorization Techniques for Recommender Systems. IEEE Computer Journal, 42, 30-37. [Google Scholar] [CrossRef] 
[22] Goyal, P. and Ferrara, E. (2017) Graph Embedding Techniques, Appli-cations, and Performance: A Survey. Knowledge Based Systems, 151, 78-94. [Google Scholar] [CrossRef] 
[23] Dong, X., Gabrilovich, E. and Heitz, G. (2014) Knowledge Vault: A Web-Scale Approach to Probabilistic Knowledge Fusion. In: Proceedings of the 20th ACM SIGKDD Confer-ence on Knowledge Discovery and Data Mining, Association for Computing Machinery, New York, 601-610. [Google Scholar] [CrossRef] 
[24] Wu, S., Tang, Y., Zhu, Y., et al. (2019) Session-Based Recom-mendation with Graph Neural Networks. In: Proceedings of the 2019 33rd AAAI Conference on Artificial Intelligence, AAAI Press, Palo Alto, 346-353. [Google Scholar] [CrossRef] 
[25] Yu, F., Zhu, Y.Q., Liu, Q., et al. (2020) TAGNN: Target Atten-tive Graph Neural Networks for Session-Based Recommendation. In: Proceedings of the 2020 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval, Association for Computing Machinery, New York, 1921-1924. [Google Scholar] [CrossRef] 
[26] 任俊伟, 曾诚, 肖丝雨, 等. 基于会话的多粒度图神经网络推荐模型[J]. 计算机应用, 2021, 41(11): 3164-3170.
[27] 刘洪伟, 高鸿铭, 陈丽, 等. 基于用户浏览行为的兴趣识别管理模型[J]. 数据分析与知识发现, 2018, 2(2): 74-85.
[28] Tan, Y.K., Xu, X. and Liu, Y. (2016) Improved Re-current Neural Networks for Session-Based Recommendations. In: DLRS 2016: Proceedings of the 1st Workshop on Deep Learning for Recommender Systems, Association for Computing Machinery, New York, 17-22. [Google Scholar] [CrossRef]