Müşteri Kaybetme Tahmini (Advanced Customer Churn Prediction Using Machine Learning)

This research addresses the role of machine learning techniques in enhancing the prediction and comprehension of customer churn within the context of the increasingly competitive e-commerce sector and rising customer expectations. The study is conducted on a comprehensive dataset comprising data from approximately 49,000 customers with 49 initial features. The feature selection process employed the mRMR method, focusing on the 15 most influential features in predicting customer churn. These selected features were then used to compare various machine learning models such as Random Forest, Support Vector Machine (SVM), Logistic Regression, and XGBoost 2.0. The performance of these models was assessed using metrics such as accuracy, recall, precision, and F1-score. Furthermore, in alignment with Explainable AI principles, SHAP values were utilized to thoroughly examine the rationale behind the models’ predictions, thereby enhancing understanding of the driving forces behind customer attrition. The findings of the research offer valuable insights to e-commerce companies, guiding them in making strategic and data-driven decisions to reduce customer loss and increase customer engagement.