A HYBRID RECOMMENDER SYSTEM FOR ONLINE PLATFORMS USING COLLABORATIVE AND CONTENT-BASED FILTERING WITH BAYESIAN HYPERPARAMETER OPTIMIZATION

Muxtarova A. I. & Artikiva M. A.


Annotatsiya

This paper presents a hybrid recommender system designed for online platforms, combining Singular Value Decomposition (SVD)-based collaborative filtering with TF-IDF-based content-based filtering. The system addresses the core limitations of single-paradigm approaches - data sparsity and the cold-start problem - through a weighted combination mechanism whose parameters are jointly optimised using Bayesian search via the Optuna library. Experiments conducted on the Amazon Reviews 2023 Video Games dataset (686,004 interactions, 89,615 users, 23,531 items, 99.5% sparsity) demonstrate that the content-based component achieves Precision@10 = 0.0014 and MRR = 0.0038, outperforming the popularity baseline across personalisation metrics. A three-tier fallback strategy handles cold-start scenarios, and the system is deployed as a REST API on Hugging Face Spaces, demonstrating real-time applicability.


Adabiyotlar

  1. G. Linden, B. Smith, and J. York, "Amazon.com recommendations: Item-to-item collaborative filtering," IEEE Internet Computing, vol. 7, no. 1, pp. 76–80, 2003.
  2. C. A. Gomez-Uribe and N. Hunt, "The Netflix recommender system: Algorithms, business value, and innovation," ACM Trans. Manag. Inf. Syst., vol. 6, no. 4, pp. 1–19, 2016.
  3. Y. Koren, R. Bell, and C. Volinsky, "Matrix factorization techniques for recommender systems," IEEE Computer, vol. 42, no. 8, pp. 30–37, 2009.
  4. P. Lops, M. de Gemmis, and G. Semeraro, "Content-based recommender systems: State of the art and trends," in Recommender Systems Handbook, Springer, 2011, pp. 73–105.
  5. R. Burke, "Hybrid recommender systems: Survey and experiments," User Modeling and User-Adapted Interaction, vol. 12, no. 4, pp. 331–370, 2002.
  6. X. He, L. Liao, H. Zhang, L. Nie, X. Hu, and T.-S. Chua, "Neural collaborative filtering," in Proc. WWW, 2017, pp. 173–182.
  7. N. Reimers and I. Gurevych, "Sentence-BERT: Sentence embeddings using Siamese BERT-networks," in Proc. EMNLP, 2019, pp. 3982–3992.
  8. T. Akiba, S. Sano, T. Yanase, T. Ohta, and M. Koyama, "Optuna: A next-generation hyperparameter optimization framework," in Proc. KDD, 2019, pp. 2623–2631.
  9. J. Hou, Z. Li, H. He, W. Xu, Y. Zhang, W. Wang, and J. McAuley, "Bridging language and items for retrieval and recommendation," arXiv:2403.03952, 2024.
  10. A. I. Schein, A. Popescul, L. H. Ungar, and D. M. Pennock, "Methods and metrics for cold-start recommendations," in Proc. SIGIR, 2002, pp. 253–260.