<![CDATA[Telemarketing campaigns are widely used by banks to promote term deposit products, yet their success rate is often low because the offers are not targeted to the right customers. This study aims to compare several machine learning models for predicting customers’ acceptance of term deposit offers, so that banks can conduct more effective and efficient campaigns. The dataset used is the Bank Marketing Dataset, which contains 45,211 customer records with demographic, socio-economic, and campaign-related attributes. The research stages include exploratory data analysis to understand the data characteristics and class imbalance, followed by data preprocessing such as handling “unknown” values, encoding categorical variables using one-hot encoding, transforming the target label into binary classes, and splitting the data into training and test sets using a stratified scheme. The models evaluated in this study are Decision Tree, Random Forest, and XGBoost, which are further optimized using Grid Search Cross-Validation. Model performance is measured using accuracy, precision, recall, and F1-score. The experimental results show that the tuned XGBoost model achieves the best performance with accuracy above 90% and stable results across different data subsets. This model can be utilized as a decision support tool to prioritize customers with a high probability of accepting term deposit offers and to improve the efficiency of telemarketing campaigns.]]>
Copyrights © 2026
