Predicting Loan Eligibility Approval Using Machine Learning Algorithms

Guangxuan Chen

2024

Abstract

The survival and profitability of financial institutions are closely related to the recipients of the loan. However, traditional loan approval methods are struggling to keep up with the diversifying and rapidly growing loan applications. In this context, machine learning techniques present a promising solution. Previous studies have only attempted limited models, while this study aims to achieve higher loan approval prediction accuracy by comprehensively comparing multiple mainstream models. This article selects a publicly available dataset from Kaggle, conducts detailed data preprocessing, and comprehensively trains eight mainstream models. The evaluation metrics used are precision, accuracy, and F1-score. Among these models, AdaBoost performed the best, achieving the highest Accuracy (84.95%) and the best F1-score (0.8957). XGBoost performed the best in terms of Precision. This study presents a more accurate method for loan approval and demonstrates the reliability of machine learning in supporting intelligent financial decision-making. This tool helps financial institutions to efficiently and fairly assess the eligibility of loan applicants, streamline the loan approval process, and promote financial inclusion.

Paper Citation

in Harvard Style

Chen G. (2024). Predicting Loan Eligibility Approval Using Machine Learning Algorithms. In Proceedings of the 1st International Conference on Data Science and Engineering - Volume 1: ICDSE; ISBN 978-989-758-690-3, SciTePress, pages 512-517. DOI: 10.5220/0012828200004547

in Bibtex Style

@conference{icdse24,
author={Guangxuan Chen},
title={Predicting Loan Eligibility Approval Using Machine Learning Algorithms},
booktitle={Proceedings of the 1st International Conference on Data Science and Engineering - Volume 1: ICDSE},
year={2024},
pages={512-517},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012828200004547},
isbn={978-989-758-690-3},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Data Science and Engineering - Volume 1: ICDSE
TI - Predicting Loan Eligibility Approval Using Machine Learning Algorithms
SN - 978-989-758-690-3
AU - Chen G.
PY - 2024
SP - 512
EP - 517
DO - 10.5220/0012828200004547
PB - SciTePress