A comparative study of customer churn prediction in telecom industry using ensemble based classifiers

Churn Prediction plays a vital role in various domains like life insurance, banking and telecom industry. With the current advancement in Machine Learning and Artificial Intelligence, Churn Prediction is more realistic and accurate. It is very much essential for early stage detection of customers who are at high risk of leaving the company or services. In this paper, Ensemble based Classifiers namely Bagging, Boosting and Random Forest were utilized for Churn Prediction in telecom industry. The Ensemble based Classifiers were compared with the well-known classifiers namely Decision Tree, Naïve Bayes Classifier and Support Vector Machine (SVM). The experimental results shows that Random Forest has less error rate, low specificity, high sensitivity and greater accuracy of 91.66% as compared to other methods.