What question did this study set out to answer?

The central aim is to develop a machine learning system to assess credit default risk for lenders.

January 20, 20260 citationsOpen Access

Credisure-Credit Lending Risk Analysis

BGBhavika Gupta RKRajeshwari Kanawat KAKhushi Alani

Key Points

The central aim is to develop a machine learning system to assess credit default risk for lenders.
Automated data ingestion and cleaning from real-world lending datasets
Feature engineering to enhance model inputs
Model training using CatBoost algorithm
Evaluation of model performance with accuracy and AUCROC metrics
Deployment via a web interface for practical use
Achieved an accuracy of 92.4% on held-out data
Attained an AUCROC of 0.97, indicating high model performance
Categorized borrower risk into four levels: 'Low', 'Moderate', 'High', 'Critical'

Abstract

In this work we present a full-lifecycle machine learning pipeline designed to forecast credit default risk, using real-world lending datasets. The system automates data ingestion, cleaning, feature engineering, model training (using CatBoost), evaluation and deployment via a web interface. We achieve accuracy of 92.4%, AUCROC 0.97 on held out data. The goal is to give lenders a reliable system to categorize borrower risk (“Low”, “Moderate”, “High”, “Critical”) and thus support more informed credit decisioning. We describe the architecture, methods, key results, practical deployment and implications for financial institutions.

问 AI

Bookmark

View Full Paper