<![CDATA[Customer churn prediction has become a critical aspect of business analytics, particularly in the automotive after-sales service industry. This study aims to develop an effective predictive model for identifying customers at risk of churn using big data analytics and machine learning techniques. The research focuses on four-wheeled vehicle after-sales services provided by Brand X, leveraging historical customer data over a seven-year period. Two machine learning algorithms Decision Tree and Random Forest were applied to classify churn behavior. Feature importance analysis was conducted to identify key variables influencing churn, including Warranty Status, Total Service Frequency, and Dissatisfaction Level. The models were evaluated using accuracy, sensitivity, specificity, confusion matrix, and feature importance metrics.. The findings suggest that integrating big data analytics with ensemble machine learning methods enhances churn prediction accuracy, enabling targeted customer retention strategies. This research contributes both academically and practically by providing a robust predictive framework for churn management in the automotive after-sales sector.]]>
Copyrights © 2025
