Predictive Analysis on Churn Customers in Business Industries using Supervised Machine Learning Algorithms and Smote

Abstract

The churn customers are the big loss  for the enterprise and business industry. A customer which are loyal and using a particular  products and services for  along time, suddenly taking a decision to change their mind set to use another products and services. If the industries or enterprises know that these customers are planning to migrate  and want to use another product and services in that situations they can use business strategies to protect them. This research work is related to a predictive model on churn customer  based on their credit score, history of buying patterns , geography location, gender specification and estimated salary with proper balance.  The proposed predictive model is based on the 10K customers data  which were collected from the different geographical location. The predictive Models Used: Logistic Regression, Support Vector Machines, Random Forests, Gradient Boosting, XGBoost  as base learners. The results of this churn predictive model is comparaed with that of the logistic regression, SVM, Gradient Boosting , Random Forest algorithms. The predictive model is based on hybrid XGboost with SMOTE which is used for management of imbalanced data and also hypertuned and is having accuracy 96%. Here, it is assured that this predictive hybrid model could be utilized in the industries which are service oriented in predicting of whether a loyal customer will churn or not.