Publications

Abstract : This paper describes the creation of a predictive machine learning model that can assist banks in determining who is eligible to apply for loans based on financial records. The proposed model is designed so that bankers can extract only those entries that have been signed for electronically, have a credit score greater than 700, and have a minimum of Rs.5000 after repaying the loan. The model is based on 1.25 percent interest rate. Artificial Neural Network, Gradient Descent, XgBoost, Random Forest, and Support Vector Machine methods yield 100, 50, 57.1428, 42.857, and 57.1428 percentages respectively. The results showed that Artificial Neural Networks outperformed other techniques in terms of accuracy, providing the best solution for the proposed model.