INTELLIGENT INFORMATION SYSTEM FOR PERSONALIZED RECOMMENDATION BASED ON USER INTERACTION HISTORY
DOI:
https://doi.org/10.32689/maup.it.2025.3.6Keywords:
recommender systems, emotional intelligence, graph neural networks, machine learning, personalization, movie industryAbstract
The article is devoted to the development and deployment of the Emotion-Aware Recommender information system, which combines machine learning methods and graph neural networks to enhance the accuracy of personalized movie recommendations.The methodology includes collecting and preprocessing data from multiple sources – user interaction history, ratings,textual reviews, and emotional annotations; constructing a heterogeneous graph with nodes “user”, “movie”, “genre”, and “emotion”; using ensemble models (XGBoost, LightGBM, CatBoost) for rating prediction; and employing a Heterogeneous GraphTransformer (HGT) to predict emotions and improve ranking.The scientific novelty of the work lies in the integration of emotional context into the recommendation process at the graph-relationship level, applying multi-task learning, and ensuring explainability via attention mechanisms and SHAPanalysis. Experimental results show that the proposed system significantly improves HR@10, NDCG@10, and Macro-F1 metricscompared to baseline models.Conclusions demonstrate that accounting for emotions increases recommendation relevance and user satisfaction, and the system has potential for adaptation in other domains such as music, literature, or educational platforms.
References
Adomavicius G., Tuzhilin A. Toward the next generation of recommender systems: A survey of the state-of-the- art and possible extensions. IEEE Transactions on Knowledge and Data Engineering. 2005. Vol. 17, No. 6. P. 734–749. URL: https://doi.org/10.1109/TKDE.2005.99
Cambria E., Poria S., Hazarika D., Kwok K. SenticNet 7: A commonsense-based sentiment and emotion lexicon for social media. Proceedings of the AAAI Conference on Artificial Intelligence. 2022. Vol. 36, No. 11. P. 12364–12371. URL: https://doi.org/10.1609/aaai.v36i11.21448
He X., Liao L., Zhang H., Nie L., Hu X., Chua T.-S. Neural collaborative filtering. Proceedings of the 26th International Conference on World Wide Web (WWW). 2017. P. 173–182. URL: https://doi.org/10.1145/3038912.3052569
Hamilton W., Ying R., Leskovec J. Inductive representation learning on large graphs. Advances in Neural Information Processing Systems (NeurIPS). 2017. Vol. 30. URL: https://proceedings.neurips.cc/paper/2017/hash/ 5dd9db5e033da9c6fb5ba83c7a7ebea9-Abstract.html
Li J., Ma J., Zhang J. Emotion-aware recommender systems: Recent advances and future directions. Information Processing & Management. 2023. Vol. 60, No. 2. P. 102115. URL: https://doi.org/10.1016/j.ipm.2022.102115
Sun J., Wang Z., Liu C., et al. Multi-modal emotion-aware recommender system with contrastive learning. Knowledge-Based Systems. 2023. Vol. 274. P. 110714. URL: https://doi.org/10.1016/j.knosys.2023.110714
Vaswani A., Shazeer N., Parmar N., et al. Attention is all you need. Advances in Neural Information Processing Systems (NeurIPS). 2017. Vol. 30. URL: https://proceedings.neurips.cc/paper/2017/hash/3f5ee243547dee91fbd053c1c4a845aa-Abstract.html
Wu L., Sun P., Hong R., et al. Graph neural networks in recommender systems: A survey. ACM Computing Surveys. 2022. Vol. 55, No. 5. P. 1–37. URL: https://doi.org/10.1145/3514226
Ying R., He R., Chen K., Eksombatchai P., Hamilton W., Leskovec J. Graph convolutional neural networks for web-scale recommender systems. Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. 2018. P. 974–983. URL: https://doi.org/10.1145/3219819.3219890
Zhang S., Yao L., Sun A., Tay Y. Deep learning based recommender system: A survey and new perspectives. ACM Computing Surveys. 2019. Vol. 52, No. 1. P. 1–38. URL: https://doi.org/10.1145/3285029
