TIGNN-RL: Enabling time-sensitive and context-aware intelligent decision-making with dynamic graphs in recommender systems and biomechanics knowledge
Abstract
Intelligent decision-making in dynamic recommender systems is crucial for capturing temporal user preferences and optimizing long-term user satisfaction. Traditional recommender systems often rely on static modeling, neglecting the temporal dynamics of user-item interactions. To address this limitation, we propose a novel framework, Temporal Interpretability Graph Neural Network with Reinforcement Learning (TIGNN-RL), which integrates dynamic graph neural networks (DGNNs) and Proximal Policy Optimization (PPO) to optimize personalized recommendations. Specifically, our method models user-item interactions as dynamic graphs and utilizes temporal interpretability modules to encode both temporal features and node-specific static features. The temporal interpretability module assigns time-aware and interactions weights to user-item, enabling more time-sensitive and explainable dynamic embeddings. This TIGNN dynamic graph sequential embedding is processed by some LSTM modules to be used as the state of the deep reinforcement learning agent and states. We take a joint approach to training, earn graph embeddings that enable better PPO policy. To evaluate the proposed framework, we conduct experiments on three benchmark datasets: Last.fm 1K, MovieLens 1M, and Amazon Product Review. Results show that TIGNN-RL outperforms state-of-the-art baselines, which use GNNs for augmenting DRL-based RS, in terms of accuracy (NDCG@K) and diversity (ILD@K@K), demonstrating its effectiveness in dynamic and interpretable recommendation scenarios. In this research, some biomechanics knowledge is integrated to further enhance the understanding and application of the proposed framework in scenarios where user behavior is influenced by physical factors.
References
1. F. Ricci, L. Rokach, B. Shapira, P. B. Kantor. Recommender systems survey. Knowledge-Based Systems; 2013. doi：10.1016/j.knosys.2011.12.005
2. C. Wang, M. Zhang, W. Ma, et al. Modeling item-specific temporal dynamics of repeat consumption for recommender systems. In: Proceedings of the World Wide Web Conference; 2019.
3. Yunzhi Tan, Min Yang, Chengming Li, Ruifeng Xu, Yating Liu. A time-aware graph neural network for session-based recommendation. IEEE Access: Practical Innovations, Open Solutions; 2020. DOI: 10.1109/ACCESS.2020.3015480
4. Yan Zhao, Chong Chen, Xiangyu Liu, Ziliang Zhao. Two-Stage Sequential Recommendation for Side Information Fusion and Long-Term and Short-Term Preferences Modeling. Journal of Intelligent Information Systems. 2022. DOI: 10.1007/s10844-021-00683-4
5. J. Tang, K. Wang, H. Liu, et al. Time-sensitive recommendation from recurrent user activities. Advances in Neural Information Processing Systems; 2015.
6. H. Yu, J. Jannach. STAMP: Short-Term Attention / Memory Priority Model for Session-Based Recommendation. In: Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining; 2018. doi：10.1145/3219819.3220003
7. Suhaim AB, Berri J. Context-Aware Recommender Systems for Social Networks: Review, Challenges and Opportunities. IEEE Access. 2021; 9: 57440–57463. doi: 10.1109/access.2021.3072165
8. Venkatachalam P, Ray S. How do context-aware artificial intelligence algorithms used in fitness recommender systems? A literature review and research agenda. International Journal of Information Management Data Insights. 2022; 2(2): 100139. doi: 10.1016/j.jjimei.2022.100139
9. Zhou S, Hudin NS. Advancing e-commerce user purchase prediction: Integration of time-series attention with event-based timestamp encoding and Graph Neural Network-Enhanced user profiling. Zhu J, ed. PLOS ONE. 2024; 19(4): e0299087. doi: 10.1371/journal.pone.0299087
10. Hou Z, Bu F, Zhou Y, et al. DyCARS: A dynamic context-aware recommendation system. Mathematical Biosciences and Engineering. 2024; 21(3): 3563–3593. doi: 10.3934/mbe.2024157
11. Ali W, Kumar J, Mawuli CB, et al. Dynamic context management in context-aware recommender systems. Computers and Electrical Engineering. 2023; 107: 108622. doi: 10.1016/j.compeleceng.2023.108622
12. Ma L, Chen Z, Fu Y, Li, Y. Heterogeneous Graph Neural Network for Multi-behavior Feature-Interaction Recommendation. In: Pimenidis E, Angelov P, Jayne C, et al. (editors). Artificial Neural Networks and Machine Learning—ICANN 2022. Springer, Cham; 2022.
13. Jianxin Chang, Chen Gao, Xiangnan He, Depeng Jin, Yong Li. Adaptive user modeling with long and short-term preferences for personalized recommendation. IJCAI; 2019. DOI: 10.24963/ijcai.2019/375
14. Jeong SY, Kim YK. Deep Learning-Based Context-Aware Recommender System Considering Change in Preference. Electronics. 2023; 12(10): 2337. doi: 10.3390/electronics12102337
15. Zhiheng Liu, Le Wu, Lei Chen, Richang Hong. Dynamic time-aware collaborative sequential recommendation with attention-based network. Knowledge and Information Systems; 2023. DOI: 10.1007/s10115-023-01857-y
16. Xiaokun Zhang, Bo Xu, Liang Yang, Hongfei Lin. Time interval-aware graph with self-attention for sequential recommendation. In: Proceedings of the 2022 5th International Conference on Algorithms, Computing and Artificial Intelligence; 2023. DOI: 10.1145/3576836.3576852
17. B. Hidasi, A. Karatzoglou, L. Baltrunas, D. Tikk. Session-based Recommendations with recurrent neural networks. doi: 10.48550/arXiv.1511.06939.
18. Wang L, Jin D. A Time-Sensitive Graph Neural Network for Session-Based New Item Recommendation. Electronics. 2024; 13: 223. doi:10.3390/electronics13010223
19. Shan G. Exploring the intersection of equipment design and human physical ability: Leveraging biomechanics, ergonomics/anthropometry, and wearable technology for enhancing human physical performance. Advanced Design Research. 2023; 1(1): 7–11. doi: 10.1016/j.ijadr.2023.04.001
20. Z. Wu, S. Pan, F. Chen, et. al. Graph Neural Networks: A Review of Methods and Applications. AI Open; 2018.
21. R. Chen, Y. Rubanova, J. Bettencourt, D. Duvenaud. Deep learning for sequential recommendation: algorithms, influential factors, and evaluations. doi:10.1145/342672
22. R. Chen, Y. Rubanova, J. Bettencourt, D. Duvenaud. Latent ODEs for irregularly-sampled time series. doi: 10.48550/arXiv.1907.03907.
Copyright (c) 2025 Author(s)
This work is licensed under a Creative Commons Attribution 4.0 International License.
Copyright on all articles published in this journal is retained by the author(s), while the author(s) grant the publisher as the original publisher to publish the article.
Articles published in this journal are licensed under a Creative Commons Attribution 4.0 International, which means they can be shared, adapted and distributed provided that the original published version is cited.
Submit a Paper
