The rapid growth of subscription-based services in the telecom industry has led to a larger subscriber base for service vendors. However, customer churn, or the loss of clients, has become a critical issue for telecom companies. This paper addresses the problem of customer churn or attrition in the telecommunication industry and performs a churn prediction model that implements Machine Learning techniques on a big data platform. The study employs key algorithms, including Logistic Regression, Decision Tree, KNN Classifier, Random Forest and Support Vector Machine (RBF), to identify potential defectors and enable companies to implement retention techniques. This paper focuses on voluntary churn and implements engineering features and selection methods to improve predictive accuracy. As markets expand and customer acquisition costs increase, effective customer retention becomes paramount, prompting service companies to invest in prediction analytics for proactive churn management. The study leverages prior research and experiments with multiple machine-learning algorithms to build a predictive model of customer churn. The utilized model uses big data and innovative feature engineering and selection methods to predict customers who are most likely to churn.