Comparing and Evaluating Machine Learning Algorithms for Predicting Customer Churn In Telecommunication Industry

A. Isabella Amali, Dr.R. Arunkumar and Dr.S. Mohan

To sustain in current in telecommunication business, it is important to find churn in customer. Customer churn prediction (CCP) is a serious issue and needs worthy predictor. Finding an effective predictor to use in large and inappropriate data of telecom industry is tedious work. In this paper, a try to compare and evaluate performance of various data mining techniques such as decision tree (DT), logistic regression (LR) support vector machine (SVM), linear classifier (LC) and Naïve Bayes (NB) on the customer churn prediction is done. An applied algorithm is evaluated for performance using few datasets. Naïve Bayes (NB) is found to be effective than other classifiers.

Volume 11 | 06-Special Issue

Pages: 170-178