基于注意力机制的用户动态兴趣推荐算法
User Dynamic Interest Recommendation Algorithm Based on Attention Mechanism

DOI: 10.12677/CSA.2022.122027, PP. 269-279

Keywords: 推荐算法，注意力机制，潜在因子模型，循环神经网络，门控循环单元
Recommendation Algorithm, Attention Mechanism, Latent Factor Model, Recurrent Neural Network, Gated Recurrent Unit

Full-Text   Cite this paper   Add to My Lib

Abstract:

传统推荐算法由于缺乏对用户兴趣的精准捕捉，会导致算法性能不佳等问题。随着深度学习在推荐系统领域应用的不断深入，针对该问题有了许多很好的解决方法。为实现对用户动态兴趣的捕捉与行为的预测，提出一种基于注意力机制的用户动态兴趣推荐算法(Attention Mechanical Recom-mendation Algorithm Based on Long Short-Term, AM_LST)。该算法通过潜在因子模型与门控循环单元，分别实现对用户长短期兴趣的动态捕捉；并使用注意力机制将用户长短期兴趣有机结合，来改善传统推荐算法因用户兴趣的动态变化而导致推荐效果下降的问题。最终通过对比实验，证明了本文提出的模型在性能上的提升。
Due to the lack of accurate capture of user interest, the traditional recommendation algorithm will result in poor performance. With the application of deep learning in the field of recommendation system, there are many good solutions to this problem. In order to realize the capture of user’s dynamic interest and behavior prediction, a user’s dynamic interest recommendation algorithm based on attention mechanism (Attention Mechanical Recommendation Algorithm Based on Long Short-Term, AM_LST) is proposed. In this algorithm, latent factor model and gated recurrent unit are used to capture the user’s long short-term interests. In addition, attention mechanism is used to organically combine users’ long and short interests to improve the problem that the traditional recommendation algorithm’s recommendation effect decreases due to the dynamic change of users’ interests. Finally, the performance improvement of the proposed model is proved through comparative experiments.

References

[1]	Rendle, S., Freudenthaler, C. and Schmidt-Thieme, L. (2010) Factorizing Personalized Markov Chains for Next-Basket Recommendation. Proceedings of the 19th International Conference on World Wide Web, Raleigh, 26-30 April 2010, 811-820. https://doi.org/10.1145/1772690.1772773
[2]	He, R. and McAuley, J. (2016) Fusing Similarity Models with Markov Chains for Sparse Sequential Recommendation. 2016 IEEE 16th International Conference on Data Mining (ICDM), Barcelona, 12-15 December 2016, 191-200. https://doi.org/10.1109/ICDM.2016.0030
[3]	Xu, C., Zhao, P., Liu, Y., Xu, J., Sheng, S., Cui, Z., et al. (2019) Recurrent Convolutional Neural Network for Sequential Recommendation. The World Wide Web Conference, San Fran-cisco, 13-17 May 2019, 3398-3404. https://doi.org/10.1145/3308558.3313408
[4]	Tang, J. and Wang, K. (2018) Personalized Top-N Sequential Recommendation via Convolutional Sequence Embedding. Proceedings of the 11th ACM International Conference on Web Search and Data Mining, Marina del Rey, 5-9 February 2018, 565-573. https://doi.org/10.1145/3159652.3159656
[5]	Zhou, G., Zhu, X., Song, C., Zhu, H., Ma, X., Yan, Y., et al. (2018) Deep Interest Network for Click-Through Rate Prediction. Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, London, 19-23 August 2018, 1059-1068. https://doi.org/10.1145/3219819.3219823
[6]	Zhou, G., Mou, N., Fan, Y., Pi, Q., Bian, W., Zhou, C., et al. (2019) Deep Interest Evolution Network for Click-Through Rate Prediction. Proceedings of the AAAI Conference on Ar-tificial Intelligence, 33, 5941-5948.
[7]	Feng, Y., Lv, F., Shen, W., Wang, M., Sun, F., Zhu, Y., et al. (2019) Deep Session Interest Network for Click-Through Rate Prediction. Proceedings of the 28th International Joint Conference on Artificial Intelligence, Macao (China), 10-16 August 2019, 2301-2307. https://doi.org/10.24963/ijcai.2019/319
[8]	Xiao, J., Ye, H., He, X., Zhang, H., Wu, F. and Chua, T.-S. (2017) Attentional Factorization Machines: Learning the Weight of Feature Interactions via Attention Networks. Proceedings of the 26th International Joint Conference on Artificial Intelligence, Melbourne, 19-25 August 2017, 3119-3125. https://doi.org/10.24963/ijcai.2017/435
[9]	王升升, 赵海燕, 陈庆奎, 曹健. 个性化推荐中的隐语义模型[J]. 小型微型计算机系统, 2016, 37(5): 881-889.
[10]	Liu, W., Wang, B. and Wang, D. (2018) Improved Latent Factor Model in Movie Recommendation System. 2018 International Conference on Intelligent Autonomous Systems (ICoIAS), Singapore, 1-3 March 2018, 101-104. https://doi.org/10.1109/ICoIAS.2018.8494074
[11]	Hidasi, B., Karatzoglou, A., Baltrunas, L. and Tikk, D. (2015) Session-Based Recommendations with Recurrent Neural Networks.
[12]	Chen, J. and Abdul, A. (2019) A Ses-sion-Based Customer Preference Learning Method by Using the Gated Recurrent Units with Attention Function. IEEE Access, 7, 17750-17759. https://doi.org/10.1109/ACCESS.2019.2895647
[13]	Cui, R., Wang, J. and Wang, Z. (2021) Multiplicative Attention Mechanism for Multi-Horizon Time Series Forecasting. 2021 International Joint Con-ference on Neural Networks (IJCNN), Shenzhen, 18-22 July 2021, 1-6. https://doi.org/10.1109/IJCNN52387.2021.9533598
[14]	谢恩宁, 何灵敏, 王修晖. 基于注意力机制的深度协同过滤模型[J]. 中国计量大学学报, 2019, 30(2): 219-225+242.
[15]	He, X., Liao, L., Zhang, H., Nie, L., Hu, X. and Chua, T.-S. (2017) Neural Collaborative Filtering. Proceedings of the 26th International Conference on World Wide Web, Perth, 3-7 April 2017, 173-182. https://doi.org/10.1145/3038912.3052569

Full-Text