Unified Graph-Temporal Recommendation Model Based on Preference Drift and Prediction
Article Sidebar
Main Article Content
Modern recommender systems operate in highly dynamic environments where user preferences evolve over time. This phenomenon, commonly referred to as preference drift, leads to performance degradation when models assume stationary behavior. In this work, we propose a drift-aware recommendation framework that models and predicts user preference evolution using graph neural networks combined with temporal sequence modeling. We represent user–item interactions as a sequence of time-dependent bipartite graphs and learn user embeddings via GraphSAGE. Temporal dynamics are captured using a recurrent neural network that predicts future user embeddings. We further incorporate Monte Carlo dropout to estimate predictive uncertainty and quantify drift magnitude as geometric displacement in latent space. Experiments on MovieLens and Yambda datasets demonstrate that the proposed method improves embedding prediction accuracy, enhances recommendation robustness under high-drift conditions, and enables reliable detection of anomalous behavioral changes. The results show that drift-aware modeling significantly mitigates performance degradation in non-stationary environments while providing actionable uncertainty signals.
Downloads
References
Kipf, T. N., & Welling, M. (2016). Semi-supervised classification with graph convolutional networks. ICLR.
Hamilton, W., Ying, Z., & Leskovec, J. (2017). Inductive representation learning on large graphs. NeurIPS.
Hidasi, B., Karatzoglou, A., Baltrunas, L., & Tikk, D. (2015). Session-based recommendations with recurrent neural networks. ICLR.
Kang, W. C., & McAuley, J. (2018). Self-attentive sequential recommendation. ICDM.
He, X., Deng, K., Wang, X., Li, Y., Zhang, Y., & Wang, M. (2020). LightGCN: Simplifying and powering graph convolution network for recommendation. SIGIR.
Rossi, E., Chamberlain, B., Frasca, F., Eynard, D., Monti, F., & Bronstein, M. (2020). Temporal graph networks for deep learning on dynamic graphs. ICML Workshop.
Liu, Y., et al. (2024). SelfGNN: Self-supervised graph neural networks for sequential recommendation. arXiv preprint.
Bifet, A., & Gavalda, R. (2007). Learning from time-changing data with adaptive windowing. SDM.
Korobov, M., et al. (2016). Dynamic embeddings for user profiling in Twitter. KDD.
Ding, H. (2025). Capturing dynamic user preferences with temporal drift modeling. MDPI Information.
Koren, Y., Bell, R., & Volinsky, C. (2009). Matrix factorization techniques for recommender systems. Computer, 42(8), 30–37.
Koren, Y. (2010). Collaborative filtering with temporal dynamics. Communications of the ACM, 53(4), 89–97.
He, X., Liao, L., Zhang, H., Nie, L., Hu, X., & Chua, T.-S. (2017). Neural collaborative filtering. WWW.
Sun, F., Liu, J., Wu, J., Pei, C., Lin, X., Ou, W., & Jiang, P. (2019). BERT4Rec: Sequential recommendation with bidirectional encoder representations from transformer. CIKM.
Wu, Z., Pan, S., Chen, F., Long, G., Zhang, C., & Yu, P. S. (2025). Graph foundation models for recommendation: A survey and future directions. ACM Computing Surveys, 57(2), 1–38.
Gal, Y., & Ghahramani, Z. (2016). Dropout as a Bayesian approximation: Representing model uncertainty in deep learning. ICML.
Liang, D., Krishnan, R. G., Hoffman, M. D., & Jebara, T. (2018). Variational autoencoders for collaborative filtering. WWW.
Wang, C., et al. (2023). Calibrated uncertainty estimation for recommender systems. WSDM.
Zhang, Y., et al. (2025). Preference drift modeling with graph recurrent networks. Data Mining and Knowledge Discovery, 39(1), 1–28.
Chen, L., Zhang, H., & Li, T. (2023). Hierarchical adaptive temporal tree for drift detection in recommender systems. IEEE Transactions on Knowledge and Data Engineering, 35(6), 6123–6137.
Zhao, P., Xie, X., Zhou, X., & Liu, Y. (2024). TDTL: A drift-aware temporal recommender system based on trust decay. Neurocomputing, 599, 128–145.
Xiao, Y., Wang, H., Zhang, Y., & Wang, J. (2024). SILSSRec: A sequential recommender system with long and short-term preferences based on contrastive learning. Electronics, 13(18), 3710.
Fan, W., et al. (2022). Incorporating dynamic user interests into sequential recommendation via hypergraph attention networks. IEEE Transactions on Knowledge and Data Engineering.
Liu, Q., et al. (2024). A comprehensive survey on graph neural networks for dynamic graphs. IEEE Transactions on Neural Networks and Learning Systems.
Li, Y., et al. (2025). Uncertainty-aware session-based recommendation with Monte Carlo dropout. Information Sciences, 680, 121–135.
Zhou, K., et al. (2020). S3-Rec: Self-supervised learning for sequential recommendation with mutual information maximization. CIKM.
Xu, Y., et al. (2024). Temporal knowledge graph reasoning for sequential recommendation. The Web Conference.
Zhang, J., et al. (2022). Multi-interest evolution modeling for sequential recommendation. KDD.
Chen, L., & Pu, P. (2024). Trust-aware recommendation with uncertainty quantification. User Modeling and User-Adapted Interaction, 34(2), 215–248.
Liang, D., et al. (2016). Bernoulli matrix factorization for implicit feedback. WWW.
This work is licensed under a Creative Commons Attribution 4.0 International License.
All articles published in our journal are licensed under CC-BY 4.0, which permits authors to retain copyright of their work. This license allows for unrestricted use, sharing, and reproduction of the articles, provided that proper credit is given to the original authors and the source.