Analyzing the Performance of Ensemble Machine Learning Algorithms for Predicting Loan Eligibility

Loans are becoming increasingly important in the banking sector, but traditional methods of assessing them based mainly on asset value and income often fall short in identifying reliable borrowers. Our research addresses this gap by using machine learning techniques to predict loan eligibility more...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:2024 9th International Conference on Communication and Electronics Systems (ICCES) S. 1362 - 1367
Hauptverfasser: C, Santhosh Kumar, S, Manishankar, Reddy, Perla Madhava, Gopal, Keertan
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 16.12.2024
Schlagworte:
Accuracy
Banking
Branch-specific assessment
Decision trees
Extra Trees algorithm
Industries
Loan default reduction
Loan eligibility prediction
Machine learning
Machine learning algorithms
Online loan application
Prediction algorithms
Reliability
Uncertainty
User interfaces
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Loans are becoming increasingly important in the banking sector, but traditional methods of assessing them based mainly on asset value and income often fall short in identifying reliable borrowers. Our research addresses this gap by using machine learning techniques to predict loan eligibility more effectively. We apply algorithms such as Decision Tree, Extra Trees, XGBoost, and LightGBM, with LightGBM achieving an optimal accuracy of 98.91%, to analyze data and produce more accurate predictions. We also introduce a new feature that allows users to input branch-specific details, enabling customized loan assessments based on different bank criteria within each branch. This feature helps users identify banks most likely to approve their loan applications. After predictions are made, the platform displays the locations of the relevant bank branches. Our system allows users to apply for loans online and provides detailed information on various loan types and branch-specific requirements. This streamlined approach reduces loan defaults, accelerates approvals, and offers clients more suitable credit options tailored to their financial circumstances.
DOI:10.1109/ICCES63552.2024.10859945