An Assessment on the machine learning approaches for credit scoring and default prediction: a case study of CABS bank

Author

Vuyo, Nyembezi

Title

An Assessment on the machine learning approaches for credit scoring and default prediction: a case study of CABS bank

Abstract

This research examines how machine learning methods can be used for credit scoring and default prediction at CABS Bank in Zimbabwe. Conventional credit evaluation techniques that mainly rely on historical financial data and set criteria often struggle to accurately predict credit risks, especially with the complexity of modern financial data. This study explores using advanced machine learning approaches to improve the precision and efficiency of assessing creditworthiness. The research involves creating and implementing machine learning models, comparing their effectiveness, and evaluating their fairness and ethical implications. The main findings demonstrate that machine learning models, like Random Forest and Neural Networks, outperform traditional methods in anticipating loan defaults, offering a more thorough and precise evaluation of credit risk. The research concludes by suggesting ways to incorporate these advanced methodologies into banking practices to enhance loan approval processes and promote wider financial inclusivity.

Date

June 2024

Publisher

BUSE

Keywords

Machine learning

Banking in AI

CABS bank

Credit scoring

Default prediction

Supervisor

Ms Hlupho