NORMA eResearch @NCI Library

Credit risk assessment of consumer loans in India using machine learning techniques

Komatineni, Venu Babu (2024) Credit risk assessment of consumer loans in India using machine learning techniques. Masters thesis, Dublin, National College of Ireland.

[thumbnail of Master of Science]
Preview
PDF (Master of Science)
Download (575kB) | Preview
[thumbnail of Configuration Manual]
Preview
PDF (Configuration Manual)
Download (374kB) | Preview

Abstract

This dissertation examines on how machine learning can improve the credit risk analysis with the emphasis on Random Forest Classifier. Credit risk analysis plays a vital role in the eligibility of loans so as to reduce risks within the financial institutions. Predictive models from statistical trails are sometimes less effective for loan outcomes prediction because of their inability to correctly analyze TCGA’s big data with many features. This research seeks to overcome these limitations through the use of the Random Forest Classifier, which is an ensemble learning algorithm that can handle complexities in the data sets and minimizes overfitting.

The research process is oriented toward cleaning, preparation, and modeling the data. The framework employed incorporates data that concerns the customer characteristics and loan features. Finally, preprocessing is done and the dataset is split into training and testing set with the help of Random Forest Classifier to predict the loan statuses. The evaluation of the model is done by accuracy, classification report and confusion matrix are obtained.

According to the study, the efficiency is about 66% implying that the results are acceptable for this type of analysis while the model’s performance is better when it comes to rejected loan predictions as compared to the approved loans. Thus, although the model offers significant information, it is not ideal. Suggested procedures which may improve prediction accuracy are feature engineering, dataset balancing, and hyperparameters optimization.

Item Type: Thesis (Masters)
Supervisors:
Name
Email
Onwuegbuche, Faithful
UNSPECIFIED
Uncontrolled Keywords: Credit Risk Assessment; Machine Learning; Random Forest Classifier; Predictive Modeling; Loan Status Prediction
Subjects: D History General and Old World > DS Asia
H Social Sciences > HG Finance > Credit. Debt. Loans.
H Social Sciences > HG Finance > Fintech
T Technology > T Technology (General) > Information Technology > Fintech
Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions: School of Computing > Master of Science in FinTech
Depositing User: Ciara O'Brien
Date Deposited: 05 Aug 2025 10:35
Last Modified: 05 Aug 2025 10:35
URI: https://norma.ncirl.ie/id/eprint/8423

Actions (login required)

View Item View Item