Customer Churn Prediction System Using Machine Learning: A Case Study ROSHAN Telecom-Afghanistan

The success of any business relies on its customers, so it is crucial for firms to prioritize customer satisfaction. Customer churn is a significant concern for companies due to increased competition, the growing importance of marketing strategies, and customers' awareness of their options. To address churn, organizations need to develop specific strategies that align with the services they offer. In the telecom sector, customer churn is especially important due to the high cost of acquiring new customers. This study aims to develop a customer churn prediction system in the telecom sector using machine learning, specifically the XGBoost algorithm. Real-time data from ROSHAN Telecom was used to predict risky customers likely to churn, enabling proactive management strategies for customer retention. The study's results show that the accuracy rate of churn prediction is 93.39 percent, demonstrating the effectiveness of machine learning techniques in predicting customer churn for the telecom company.