Bank Term Deposit Subscription Recommendation Using Cross-validated Neural Network

Term deposit can accelerate the financial field by enhancing the profit from both the bank and customer's perspective. The term deposit subscription is often influenced by the bank's campaign efforts as well as the customer background details. If customers' subscription tendency is identified at an early stage, the bank can modify its underlying strategy to attract more customers. In this context, the current study has focused to identify term likelihood prediction from the customer's perspective. To address this study, machine learning-based approaches have been applied to predict term deposit investment possibilities in advance. A popular machine learning-based method, Neural network along with stratified 10-fold cross-validation methodology is proposed as the predictive model in this study. To assess the efficiency of this model, other benchmark classifiers such as k-Nearest Neighbor (k-NN), Decision tree classifier (DT), and Multi-layer perceptron classifier (MLP) are also implemented and compared. This comparative study has concluded that the proposed model provides significant prediction results over other baseline models with an accuracy of 88.32% and MSE of 0.1168.