Improved Decision Support System for Personal Loan Eligibility Using Artificial Neural Networks |
||
 |
 |
|
| © 2024 by IJCTT Journal | ||
| Volume-72 Issue-7 |
||
| Year of Publication : 2024 | ||
| Authors : Kevin Macwan | ||
| DOI : 10.14445/22312803/IJCTT-V72I7P119 | ||
How to Cite?
Kevin Macwan, "Improved Decision Support System for Personal Loan Eligibility Using Artificial Neural Networks," International Journal of Computer Trends and Technology, vol. 72, no. 7, pp.138-155, 2024. Crossref, https://doi.org/10.14445/22312803/IJCTT-V72I7P119
Abstract
This paper comprehensively evaluates various machine learning models applied to four distinct datasets, emphasizing their performance in binary classification tasks. We employed multiple algorithms, including Logistic Regression, Random Forest, XGBoost, SVM, KNN, Decision Tree, LSTM, CNN, DNN, and Radial Basis Function Network (RBFN), to compare their effectiveness using metrics such as AUC-ROC, precision, recall, and overall accuracy. Home Loan Dataset results highlighted the variations in model performance, with the highest AUC value being 84% and the overall accuracy ranging from 73% to 100%. XGBoost and Decision Tree models achieved 100% accuracy, underscoring their robustness in this context. Lending Club Loan Data demonstrated stark differences in model efficacy, with AUC values varying from 50% to 100%. Here, Random Forest, XGBoost, and Decision Tree models consistently achieved perfect classification accuracy, indicating their superior handling of this dataset. Loan Default Prediction Dataset involved a more challenging classification task, reflected in lower AUC values, with the highest being 77%. The overall accuracy was around 92%, with Logistic Regression and Random Forest models showing relatively balanced performance. Bank Loan Default Dataset explored the impact of logistic regression, Random Forest, XGBoost, Decision Tree, KNN, LSTM, CNN, DNN, and RBFN models, achieving varying degrees of success. Random Forest and XGBoost again proved to be the top performers, achieving perfect accuracy, while other models like CNN and LSTM displayed limitations in specificity and recall. This study underscores the importance of selecting appropriate machine learning models based on dataset characteristics and desired performance metrics. The comparative analysis herein aims to guide practitioners in choosing the most effective algorithms for their classification challenges, ultimately enhancing data observability and decision-making processes with AI and LLMs.
Keywords
Machine Learning, Classification, Ensemble methods, Model performance, Data observability.
Reference
[1] Afrah Khan et al., “Loan Approval Prediction Model: A Comparative Analysis,” Advances and Applications in Mathematical Sciences, vol. 20, no. 3, pp. 427-435, 2021.
[Google Scholar] [Publisher Link]
[2] Hua Li et al., “XGBoost Model and its Application to Personal Credit Evaluation,” IEEE Intelligent Systems, vol. 35 no. 3, pp. 52-61, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[3] Xiong Zhong, and Sheng Zhou, “Retracted Article: Risk Analysis Method of Bank Microfinance Based on Multiple Genetic Artificial Neural Networks,” Neural Computing and Applications, vol. 32, pp. 5367-5377, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[4] S Sreesouthry et al., “Loan Prediction Using Logistic Regression in Machine Learning,” Annals of the Romanian Society for Cell Biology, vol. 25, no. 4, pp. 2790-2794, 2021.
[Google Scholar] [Publisher Link]
[5] Min Sue Park et al., “Explainability of Machine Learning Models for Bankruptcy Prediction,” IEEE Access, vol. 9, pp. 124887-124899, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[6] Abhiroop Sarkar, “Machine Learning Techniques for Recognizing the Loan Eligibility,” International Research Journal of Modernization in Engineering Technology and Science, vol. 3, no. 12, pp. 1135-1142, 2021.
[Google Scholar] [Publisher Link]
[7] Yingli Wu et al., “The Analysis of Credit Risks in Agricultural Supply Chain Finance Assessment Model Based on Genetic Algorithm and Backpropagation Neural Network,” Computational Economics, vol. 60, pp. 1269-1292, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[8] Joseph Bamidele Awotunde et al., “Artificial Intelligence Based System for Bank Loan Fraud Prediction,” 21st International Conference on Hybrid Intelligent Systems, pp. 463-472, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[9] Syed Emad Azhar Ali et al, “Predicting Delinquency on Mortgage Loans: An Exhaustive Parametric Comparison of Machine Learning Techniques,” International Journal of Industrial Engineering and Management, vol. 12, no. 1, pp. 1-13, 2021.
[Google Scholar] [Publisher Link]
[10] Marion O Adebiyi et al., “Secured Loan Prediction System Using Artificial Neural Network,” Journal of Engineering Science and Technology, vol. 17, no. 2, pp. 854-873, 2022.
[Google Scholar] [Publisher Link]
[11] Ugochukwu E. Orji et al., “Machine Learning Models for Predicting Bank Loan Eligibility,” 2022 IEEE Nigeria 4th International Conference on Disruptive Technologies for Sustainable Development (NIGERCON), Lagos, Nigeria, pp. 1-5, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[12] An-Hsing Chang et al., “Machine Learning and Artificial Neural Networks to Construct P2P Lending Credit-Scoring Model: A Case Using Lending Club Data,” Quantitative Finance and Economics, vol. 6, no. 2, pp. 303-325, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[13] S. Varshaa Sai Sripriya, Sai Divya Santoshi Varrey, and M Venkateshkumar, “Predictive Model to Compute Eligibility Test for Loans,” 2022 IEEE Industrial Electronics and Applications Conference (IEACon), pp. 185-190, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[14] Tianhui Wang, Renjing Liu, and Guohua Qi, “Multi-Classification Assessment of Bank Personal Credit Risk Based on Multi-Source Information Fusion,” Expert Systems with Applications, vol. 191, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[15] Theodore E Simos, Vasilios N Katsikis, and Spyridon D Mourtas, “A Multi-Input with Multi-Function Activated Weights and Structure Determination Neuronet for Classification Problems and Applications in Firm Fraud and Loan Approval,” Applied Soft Computing, vol. 127, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[16] Zhuo Chen et al., “An Artificial Neural Network-Based Intelligent Prediction Model for Financial Credit Default Behaviors,” Journal of Circuits, Systems and Computers, vol. 32, no. 10, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[17] G.L. Infant Cyril, and J.P. Ananth, “Deep Learning Based Loan Eligibility Prediction with Social Border Collie Optimization,” Kybernetes, vol. 52, no. 8, pp. 2847-2867, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[18] Sergio Genovesi et al., “Standardizing Fairness-Evaluation Procedures: Interdisciplinary Insights on Machine Learning Algorithms in Creditworthiness Assessments for Small Personal Loans,” AI and Ethics, vol. 4, pp. 537-553, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[19] Robert H Mnookin, and Lewis Kornhauser, “Bargaining in the Shadow of the Law: The Case of Divorce,” The Yale Law Journal, vol. 88, no. 5, pp. 950-997, 1979.
[CrossRef] [Google Scholar] [Publisher Link]
[20] Joel Greenberg, Of Prairie, Woods, and Water: Two Centuries of Chicago Nature Writing, University of Chicago Press, pp. 1-402, 2008.
[Google Scholar] [Publisher Link]
[21] Glenn Gould, The Glenn Gould Reader, Knopf, pp. 1-475, 1984.
[Google Scholar] [Publisher Link]
[22] William J. Novak, “The Myth of the ‘Weak’ American State,” The American Historical Review, vol. 113, no. 3, pp. 752-772, 2008.
[CrossRef] [Google Scholar] [Publisher Link]
[23] Geoffrey C. Ward, and Ken Burns, The War: An Intimate History, 1941-1945, Knopf, pp. 1-453, 2007.
[Google Scholar] [Publisher Link]
[24] OECD. Publishing, International Energy Agency, International Engery Agency, Energy Balances of Non-OECD Countries 2010, OECD/IEA, pp. 1-554, 2010.
[Google Scholar] [Publisher Link]
[25] International Energy Agency, Energy Balances of OECD Countries 2010, OECD, pp. 1-295, 2010. [Google Scholar] [Publisher Link]
[26] Sergey Paltsev et al., “The Cost of Climate Policy in the United States,” Energy Economics, vol. 31, pp. s235-s243, 2009.
[CrossRef] [Google Scholar] [Publisher Link]