Intelligent Banking Decision Support System Using Ensemble Machine Learning Techniques
DOI:
https://doi.org/10.64751/Abstract
Loan approval assessment depends on applicant information such as income, requested loan amount, repayment period, credit history and available assets. Manual assessment can be time-consuming, while fixed rules may not represent interactions among these attributes. This paper presents a Loan Approval Prediction prototype that integrates a machine-learning backend with a browser-based interface. The frontend is implemented using React and Vite, and the backend uses Python and scikit-learn. The visible training code includes Logistic Regression, Support Vector Classifier (SVC) and Random Forest as candidate classifiers, together with preprocessing components and feature engineering. Two derived attributes are calculated: loan-to-income ratio and total asset value. A trained pipeline is persisted using Joblib and is accessed through a locally running backend service. The interface accepts eleven applicant attributes and displays a decision, a probability value and the model label. In the supplied demonstration, the display shows “Not Approved,” “Approval probability: 69.5%,” and “random_forest.” This single example demonstrates application flow, not overall predictive accuracy. Since the available project evidence does not include verified dataset statistics or numerical test metrics, no accuracy or comparative-performance claim is made.
Published
Issue
Section
License

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.






