Моделювання вибору користувача в умовах обмежень холодного старту рекомендаційної системи

Abstract

This study explores modeling user choices in recommender systems, focusing on cold start scenarios and leveraging distributed architectures. Investigating modern approaches to recommendation systems, the research analyzes methods for predicting user interests and examines the impact of cold start challenges. It delves into defining key modeling elements and evaluating high-performance systems based on distributed architectures, highlighting their speed and scalability benefits. The findings contribute to a nuanced understanding of user choice modeling and suggest potential directions for future research in recommender system development.