This article is devoted to the theoretical foundations and modern approaches to the development of recommender systems in e-commerce. It examines system types – non-personalized, personalized, collaborative and content-based filtering, knowledge-based filtering, and hybrid solutions. Special attention is given to the “cold start” problem, methods for evaluating system performance (MAE, MSE, RMSE, Precision, Recall, F1-measure), as well as the advantages and disadvantages of recommender systems for both businesses and users.
Recommender systems, e-commerce, personalization, collaborative filtering, content-based filtering, hybrid systems, cold start, performance evaluation, MAE, RMSE, Precision, Recall, F1-measure
1. Rikardo B.A., Nil'sen A. Sovremennye podhody k postroeniyu rekomendatel'nyh sistem // Informacionnye tehnologii i analiz dannyh. – 2023. – № 4. – S. 45-58.
2. Gusev I.V. Mashinnoe obuchenie v sistemah personalizacii internet-kommercii // Elektronnaya kommerciya i cifrovye platformy. – 2024. – T. 12, № 2. – S. 33-47.
3. Kuznecov A.S., Smirnova E.Yu. Analiz effektivnosti rekomendatel'nyh sistem v onlayn-torgovle // Vestnik prikladnoy informatiki. – 2023. – № 1. – S. 19-28.
4. Davydova N.P. Primenenie metodov kollaborativnoy fil'tracii v cifrovom marketinge // Informacionnye tehnologii i programmirovanie. – 2024. URL: https://cyberleninka.ru (data obrascheniya: 10.10.2025).
5. Voronov M.E., Levina T.S. Rekomendatel'nye sistemy kak instrument povysheniya konversii v elektronnoy kommercii // Sovremennye tehnologii v ekonomike i biznese. – 2023. – № 6. – S. 57-66.
6. Goh H.A. Deep Learning Approaches for E-Commerce Recommendation Systems // International Journal of Computational Intelligence Systems. – Springer, 2023.
7. Nadukuda N. AI-Driven Personalization in E-Commerce: Techniques and Trends // International Journal of Artificial Intelligence & Applications (IJAIAP). – 2023. URL: https://iaeme.com (data obrascheniya: 10.10.2025).
