Enhanced Prediction Model for Customer Churn in Telecommunication Using EMOTE

Customer churn is the term that refers to the customers who are in threat to leave the company. A growing number of such customers are becoming critical for the telecommunication sector, and the telecom sector is also in situation to retain them to avoid the revenue loss. Prediction of such behavior is very essential for the telecom sector, and classifiers proved to be the most effective one for the same. A well-balanced data set is a vital resource for the classifiers to yield the best prediction. All existing classifiers tend to perform poorly on imbalanced data set. An imbalanced data set is the one, where the classification attribute is not evenly distributed. Like the other real-time applications, the telecommunication churn application also has the class imbalance problem. So it is extremely vital to go for fine-balanced data set for classification. In this paper, an empirical method enhanced classifier for telecommunication churn analysis model (EC_for_TELECAM) using enhanced minority oversampling technique (EMOTE) has been proposed to improve the performance of the classifier for customer churn analysis in telecom data set. To evaluate the proposed method, experiments were done with various data sets. The experimental study shows that the proposed method is able to produce well-balanced data set to improve the performance of the classifier and to produce the best prediction model.