Predictive Analysis of Bank Marketing Data for Customer Response

This research analyzes a bank's telemarketing campaign data to identify key factors influencing a customer's response. Using a dataset containing demographic information, contact history, and campaign details, the study explores the relationships between various attributes and the likelihood of deposit acceptance. We performed exploratory data analysis (EDA) to visualize trends, revealing that age, job type, marital status, and education significantly affect a customer's propensity to subscribe. Furthermore, we found a strong positive correlation between call duration and deposit acceptance. The project applies several supervised machine learning models, including Logistic Regression, K-Nearest Neighbors, Support Vector Machine (SVC), Decision Tree, Random Forest, and XGBoost, to predict the outcome of a campaign. The models were evaluated using accuracy and confusion matrices, with XGBoost and Random Forest classifiers achieving the highest accuracy, demonstrating the effectiveness of ensemble learning for this predictive task. The findings provide actionable insights for banks to optimize their marketing strategies, enabling them to target potential customers more effectively and increase campaign success rates.