• Vansh Choubey
• Prerna Gupta
• Aditya Singh
• Abhishek Ambadwar
• Himanshu Taiwade

Credit risk analysis is becoming more and more essential in this day and age as bank credit risk is a significant challenge in modern financial transactions and the ability to identify qualified credit card holders among a large number of applicants. In the past it was done by screening each applicants credit history which in turn increased manual labour and time consuming. Although credit scoring models were build they often struggled with complex and non-linear relationships and imbalanced datasets leading to inaccurate predictions. In this study we used the public dataset (name of the dataset) available on Kaggle to explore the application of Light Gradient Boosting Machine (Light-GBM) and Synthetic Minority Oversampling Technique (SMOTE) for credit Risk Analysis and also perform a comparison between various other Machine learning methods. These models are trained and compared based on the Accuracy, Precision, Recall, F1-score, and AUC-ROC to find out the effectiveness of Light-GBM. Results shows that Light-GBM combined with Smote significantly enhances the accuracy of the model and reducing the bias present in the minority class and outperforming the base line models. The experiments thus demonstrate benefits of combining gradient bosting with oversampling techniques for improved Credit Risk Analysis.

Keywords: Credit Risk Analysis, LightGBM, SMOTE, Machine Learning, Class Imbalance, Ensemble Learning, Financial Risk Prediction, AUC-ROC, Predictive Modeling.

"Enhanced Credit Risk Analysis Using Light-GBM and SMOTE: A Comparative Study with Traditional Classifiers", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 3, page no.b418-b422, March-2025, Available :http://www.jetir.org/papers/JETIR2503145.pdf

"Enhanced Credit Risk Analysis Using Light-GBM and SMOTE: A Comparative Study with Traditional Classifiers", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 3, page no. ppb418-b422, March-2025, Available at : http://www.jetir.org/papers/JETIR2503145.pdf