Гібридний метод персоналізації рекомендацій на основі поведінкових даних користувачів

Abstract

The report considers a hybrid method for personalizing recommendations based on user behavioral data in e-commerce systems. The proposed approach combines collaborative filtering, content-based filtering and behavioral data analysis to improve the relevance of recommendations. Special attention is paid to the use of implicit user interaction signals and the dynamic adjustment of algorithm weights depending on the stage of the IT project. The results demonstrate the potential of the hybrid approach to increase recommendation accuracy and improve user experience in digital commerce platforms.