Abstract
Customer churn presents a critical challenge for e-commerce businesses, directly impacting revenue and limiting long-term growth potential. Proactively identifying customers at risk of leaving is essential for implementing targeted retention strategies. This survey examines the development and application of predictive models for churn detection by analyzing various aspects of customer behavior and purchasing patterns. Leveraging a comprehensive dataset that includes transaction history, browsing behavior, and engagement metrics, we explore a range of machine learning techniques, including logistic regression, decision trees, random forests, and neural networks, for building effective churn prediction models.