Machine Learning-Driven Evaluation of Credit Risk Models and Provisioning Practices in Indian Banks

View at publisher Summarise with VIRAT
Authors: Tirlangi Kiran Kumar, D Srinivas
Cited by: 0
References: 24

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Banking in India is experiencing major problems in terms of credit risks management that can create an immense impact on the stability of the finances. The inability to capture more complex, non-linear relationships and moving factors which affect creditworthiness has often been a limitation to traditional credit risk models. This paper investigates how machine learning methods are applicable in assessing credit risk models and how it can be used to improve on practices of providing services in Indian banks. An evaluation of different machine learning models, including decision trees, random forests and neural networks is made to evaluate the predictive performance of the models in estimating credit risk. The paper further examines how the existing practice of provisioning is consistent with the current regulations and how machine learning can enhance them. The results indicate that machine learning models can better predict and provide more accurate predictions than conventional risk assessment tools and have more prudent provisioning plans. The study outlines the possibility of machine learning to enhance credit risk management systems of the banks of India to build stronger financial systems.

Fields of Study & Subjects

Financial Distress and Bankruptcy Prediction Banking stability regulation efficiency Banking Sector Performance and Management

Publication details

Journal / source
International Conference on Communication, Computing and Electronics Systems (ICCCES)
Pages
1094–1101
Publication date
21 January 2026
Language
EN