Customer churn prediction using data mining approach

Customer churn prediction is becoming one of the most important concerns to global organizations generally and in the telecommunications field specifically. For this purpose, a comparative study was conducted upon three machine learning predictive classification models applied to a dataset to predict customer churn. The three models namely; Decision Trees (DT), Naïve Bayes (NB) and Rule Induction performance was evaluated to specify the best performance by several measures such as accuracy, precision, recall, F-measure and Area Under Cover (AUC).