Bank Customer Identification for Targeted Marketing and Revenue Optimisation: A Comparative Analysis of Predictive Models

This research leads a transformative shift in banking by using machine learning to identify high value customers, crucial for targeted marketing. In the digital era, strategic customer targeting is vital for improved financial performance. Leveraging a comprehensive dataset, our study deploys various classifiers like Logistic Regression, Random Forest, KNN, Naive A, Gradient Boosting, and SVM. Through meticulous feature selection and hyperparameter tuning, SVM Hypertuned emerges as optimal classifier, showcasing superior precision, recall and F1-score. The implementation includes classifier initialisation, preprocessing, and training with a dedicated Decile analysis for nuanced insights. The research contributes a comparative analysis of classifiers, enhancing methodology in feature selection and hyperparameter tuning. With a focus on replacing manual forecasting, our motivation is to empower banks with an efficient machine learning based framework for customer identification, fostering informed decision making, strategic customer targeting, and improved financial performance. Addressing the critical challenge of efficiently identifying potential customers, the study highlights scenarios of missed opportunities in customer targeting. The proposed machine learning framework is imperative for insuring the sector's resilience and success in evolving digital landscape.