Bank Direct Marketing Campaign Success Prediction

Abstract

Nowadays, there are more new and innovative bank marketing strategies available that contribute to customer engagement and acquisition. Banks must launch marketing campaigns that can successfully attract customers from other competitors. Therefore, the purpose of the research is to apply Machine Learning techniques in predicting the success of a bank’s direct marketing campaign using supervised classification algorithms. Six algorithms are implemented which are Logistic Regression (LR), Support Vector Machine (SVM), Decision Tree (DT), Random Forest (RF), Gaussian Naïve Bayes (GNB) and Extreme Gradient Boosting (XGBoost). In this paper, feature selection is performed, and 8 selected features are used to make the prediction. The experimental result shows that the Random Forest model has the highest accuracy and ROC AUC score which makes it the champion model in this paper. Still, these results may vary according to the nature of data preprocessing and algorithms implemented.