Abstract: By obtaining capital and making loans, banks' core business model relies on financial intermediation (mortgage, real estate, and consumer and company’s loans). The latter is the biggest contributor to credit risk and is made up of two key components: loan approval and fraud. We will concentrate on loan approval using machine learning models in this post. Although banks can sell various items in our financial system, their main source of income comes from lines of credit. Therefore, they can benefit from interest on the credited loan. A bank's profit or loss is largely determined by the loans it makes, i.e. whether its customers repay the loans. Banks can reduce their non-performing assets by predicting defaults. This highlights the importance of studying this phenomenon. Based on previous research from this time, there are several ways to study the problem of default prevention. But since accurate predictions are key to maximizing returns, understanding how the different methods of work.

Keywords: Big Data, Machine Learning, Python, Logistic Regression, SVM, Decision Tree, Naive Bayes, Loan Prediction.

PDF | DOI: 10.17148/IJARCCE.2023.12316

Open chat
Chat with IJARCCE