Comparison of Naïve Bayes and C45 Methods on Predicted Deposit Customers

Abstract

Determining product customers in the banking world in marketing products such as time deposit products is a problem. This study aims to compare the methods in developing a predictive model for time deposit subscriptions with direct marketing from Portuguese banking institutions. This marketing is based on phone calls, by entering contact information data made to the client to get a "yes" or "no" answer. The comparison prediction model was built using the Naïve Bayes method and the C45 method. The results showed that the model with the highest value was built using the Naive Bayes method, which was 63.77%. The model built using the Naïve Bayes method can also link between attributes to be more accurate and make the data simple and efficient.