Credit Risk Analysis and Prediction

View Sample PDF

Author(s): Akhil Raman Sarepalli (Vellore Institue of Technology, India)and Maheshwari S (Vellore Institue of Technology, India)
Copyright: 2023
Pages: 28
Source title: Scalable and Distributed Machine Learning and Deep Learning Patterns
Source Author(s)/Editor(s): J. Joshua Thomas (UOW Malaysia KDU Penang University College, Malaysia), S. Harini (Vellore Institute of Technology, India)and V. Pattabiraman (Vellore Institute of Technology, India)
DOI: 10.4018/978-1-6684-9804-0.ch012

Keywords: Artificial Intelligence / Engineering Science Reference / Knowledge Management and Library Science / Machine Learning

Purchase

View Credit Risk Analysis and Prediction on the publisher's website for pricing and purchasing information.

Abstract

Credit scoring is used to divide applicants into two groups: those with good credit and those with bad credit. When a bank gets a loan request, borrowers with strong credit have a high likelihood of repaying debt. The likelihood of default is higher for applicants with bad credit. The profitability of financial organisations depends on the accuracy of credit scoring. Financial institutions will experience less of a loss if their credit scoring of applicants with poor credit is even 1 percent more accurate. This study seeks to solve this categorization issue by examining the risk of granting a loan to the applicant using the applicant's socioeconomic and demographic attributes from German credit data. In terms of overall accuracy, the authors evaluated the efficiency of several ML techniques like decision tree, logistic regression model, neural network, SVM, as well as random forest. The authors compared and evaluated several models for the model optimization process, integrating the impacts of balancing the AUC (area under the ROC curve)and accuracy values.

The IRMA Community

Research IRM

Credit Risk Analysis and Prediction

Purchase

Abstract

Related Content

IRMA Sponsors