Аналіз даних у веб-магазині для продажу комп’ютерних ігор: методи персоналізації та прогнозування

Abstract

This research is focused on creating a data analytics system for an online computer game store that includes personalized recommendations, forecasting sales trends, and detecting suspicious activity. The project aims to improve customer service, prevent fraudulent behavior, and increase sales efficiency using collaborative filtering and the ARIMA model. The system can predict user preferences, forecast demand for different games, and detect unusual patterns, providing both convenience for customers and valuable information for the store. In the future, these methods can be expanded to take into account social media factors and advertising campaigns, making the platform even more efficient.