Prediction of Potential Customers for Term Deposit: Analysis Using Bagging with Bank Marketing Data

The global COVID-19 pandemic triggered a severe economic downturn, impacting economies and financial systems worldwide. Lockdowns and travel restrictions disrupted supply chains, closed businesses, and led to job losses. This emphasized the need for prudent financial planning, with term deposits emerging as effective risk mitigation tools. In the realm of marketing, direct communication methods have replaced traditional intermediaries. Leveraging data from a Portuguese bank, this study analyzes the accuracy of the predictions of consumers who have a high chance of taking the term deposit. By employing machine learning algorithms, including Support Vector Machine (SVM), Gaussian Naïve Bayes, Decision Tree, Bagging, Light Gradient Boosting (GBM), bagging algorithm, and Extreme Gradient Boosting (XgBoost), to classify and predict potential customers. The findings highlight the optimal performance of bagging algorithms in predicting customer behavior, enhancing marketing strategies.