Performance Analysis of Machine Learning Algorithm for the Credit Risk Analysis in the Banking Sector

Abstract

The banking sector has advanced in recent years. Thus, there is an increase in the demand for bank loans. The bank must distribute and sell each of the limited number of available slots to a select group of people. As a result, a usual stage is to identify who will be unable to return the loan and who will prove to be a more trustworthy option to the bank. In order to save the bank time and costs, in the proposed paper machine learning based approach is introduced to reduce the risk involved with finding the safe individual. In order to decide whether or not to grant someone a loan, this paper presents a method of loan approval based on predetermined criteria. The machine learning model for credit approval was implemented using logistic regression, XG Boost, random forest and naïve bayes model. The experimental results indicates that logistic regression model is more accurate for the credit risk analysis.