Paper 2: Design and Implementation of an Intelligent Loan Eligibility System Using Machine Learning Techniques
Authors : Ayush Kashyap, Lucky Mishra, Ashutosh Mishra and Dr. Peeyush Kumar Pathak
Doi: https://doi.org/10.63920/tjths.42002
Abstract
Machine learning (ML) algorithms can bring revolution in the research field in almost all areas. Processes in numerous industries, including finance, real estate, security, and genomics, are being transformed by machine learning (ML) algorithms. One of the major impediments in the banking sector is the loan approval process. Modern tools like ML models help accelerate, streamline, and increase the precision of loan approval procedures. It will benefit both the client and the bank in terms of time and manpower required for loan eligibility prediction. The entire work is centered on a classification problem and is a form of supervised learning in which it is important to determine whether the loan will be approved or not. Also, it is a predictive modeling problem where a class label is predicted from the input data for a specific sample of input data. In this work, we deployed various ML algorithms to identify the loan approval status and compare the performance of implemented models. The implemented models will attempt to predict our target column on the test dataset using information from the loan eligibility prediction dataset obtained from Kaggle, which includes features like loan amount, number of dependents, and education. The parameters like accuracy, confusion matrix, ROC curve, and precision are measured for specific models whose performance is significant.
