Design and Creation of a Neural Network for the Classification of Bank Loan Applicants into Suitable and Non-Suitable Candidates

Abstract

In the contemporary digital era, financial institutions are actively seeking ways to enhance and streamline their decision-making processes, particularly in critical areas like the approval of bank loans. Traditionally, this procedure has heavily relied on the manual examination of applicant information against predetermined criteria. However, the advent of artificial intelligence and machine learning has ushered in a new era, where neural networks are emerging as a promising tool for intelligent decision-making, aiming to significantly enhance the precision and efficiency of these crucial determinations. The system is developed using supervised artificial neural networks. This system is trained with 8 input variables to make predictions about the feasibility of granting a loan to a user. The goal is to determine whether the application is favorable, or, on the contrary, if it presents risks or disadvantages that could make granting a loan unfavorable. It is worth noting that the system has shown good performance in terms of favorable predictions.