ENHANCING LOAN APPROVAL DECISION-MAKING - AN INTERPRETABLE MACHINE LEARNING APPROACH USING LIGHTGBM FOR DIGITAL ECONOMY DEVELOPMENT

Aug 2, 2026

Upload by Maeve

This study enhances loan approval decision-making in the digital economy by introducing an interpretable machine learning framework. The research investigates how interpretability can improve both accuracy and transparency of credit decisions. LightGBM is used for loan approval classification, tuned with Random Search and assessed via 10-fold cross-validation. SHAP values are integrated to explain model outputs, improving interpretability. Compared with baseline methods, the LightGBM model achieves 98.13% accuracy and high precision, recall, and F1-score, supporting reliable and explainable lending.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages12

Reading Time30 min

AccessFree to download

Abstract

Keywords

Introduction

Background of loan approval

Traditional assessment limitations

Machine learning and LightGBM

Need for transparency and interpretability

Malaysian Journal of Computing, 9 (1): 1734-1745, 2024 Copyright © UiTM Press eISSN: 2600-8238 ENHANCING LOAN APPROVAL DECISION-MAKING: AN INTERPRETABLE MACHINE LEARNING APPROACH USING LIGHTGBM FOR DIGITAL ECONOMY DEVELOPMENT Teuku Rizky Noviandy 1*, Ghalieb Mutig Idroes2 and Irsan Hardi3 1*Department of Informatics, Faculty of Mathematics and Natural Sciences, Universitas Syiah Kuala, Banda Aceh 23111, Indonesia 2Energy and Green Economics Unit, Graha Primera Saintifika, Aceh Besar 23371, Indonesia 3Economic Modeling and Data Analytics Unit, Graha Primera Saintifika, Aceh Besar 23371, Indonesia 1*[trizkynoviandy@gmail.com](trizkynoviandy@gmail.com), [2](2ghaliebidroes@outlook.com3irsan.hardi@gmail.com)[ghaliebidroes@outlook.com](2ghaliebidroes@outlook.com3irsan.hardi@gmail.com)[3](2ghaliebidroes@outlook.com3irsan.hardi@gmail.com)[irsan.hardi@gmail.com](2ghaliebidroes@outlook.com3irsan.hardi@gmail.com) ABSTRACT This study aims to enhance loan approval decision-making in the digital economy using an interpretable machine learning approach. The primary research question investigates how integrating an interpretable machine learning approach can improve the accuracy and transparency of loan approval processes. We employed LightGBM, a gradient-boosting framework for loan approval classification, optimized via Random Search hyperparameter tuning and validated using 10-fold cross-validation. We incorporated the Shapley Additive exPlanations (SHAP) framework to address the challenge of interpretability in machine learning. The LightGBM model outperformed conventional algorithms (Decision Tree, Random Forest, AdaBoost, and Extra Trees) in accuracy (98.13%), precision (97. 78%), recall (97.17%), and F1-score (97.48%). The study demonstrates that using an interpretable machine learning approach with LightGBM and SHAP can significantly improve the accuracy and transparency of loan approval decisions. This method offers a promising avenue for financial institutions to enhance their loan approval mechanisms, ensuring more reliable, efficient, and transparent decision-making in the digital economy. The study also underscores the importance of interpretability in deploying machine learning solutions in sectors with significant socioeconomic impacts. Keywords: Artificial Intelligence, Light Gradient Boosting Machine, Machine Learning, SHAP Received for review: 13-03-2024; Accepted: 27-03-2024; Published: 01-04-2024 DOI: 10.24191/mjoc.v9i1.25691 1. Introduction Loans are pivotal in facilitating financial transactions and enabling individuals and businesses to realize their aspirations (Kariv & Coleman, 2015) . Whether for purchasing a home, expanding a business, or covering unexpected expenses, loans are integral to economic growth and personal advancement (Makinde, 2016; Saiti & Trenovski, 2022). However, the process of This is an open access article under the CC BY-SA license ([https://creativecommons.org/licenses/by-sa/3.0/](https://creativecommons.org/licenses/by-sa/3.0/)). Noviandy et al., Malaysian Journal of Computing, 9 (1): 1734-1745, 2024 approving loans involves intricate decision-making, where financial institutions evaluate numerous factors to determine the creditworthiness of applicants. As the demand for loans continues to rise, the need for effective and efficient loan approval mechanisms becomes increasingly crucial (Dansana et al., 2023) . Traditionally, the loan approval process has been characterized by manual assessment methods that rely heavily on historical data and rigid criteria (Tchakoute Tchuigoua, 2018) . These traditional approaches often struggle to adapt to the evolving landscape of financial dynamics, resulting in inefficiencies and suboptimal decision-making. The limitations of these methods become particularly evident when faced with complex and dynamic economic conditions, leading to delays, inaccuracies, and, sometimes, overlooking potentially creditworthy applicants (

FAQs

What is the main goal of the study on loan approval?

To improve loan approval decision-making in the digital economy by using an interpretable machine learning approach that increases accuracy and transparency.

Which model and explanation method are used in the research?

The study uses LightGBM for classification and applies SHAP (Shapley Additive exPlanations) to provide interpretability for decision rationales.

How is the LightGBM model validated and tuned?

The LightGBM model is optimized using Random Search hyperparameter tuning and validated with 10-fold cross-validation.

Aug 2, 2026

Upload by Maeve