top of page

Machine Learning for Credit Scoring


This report from the Bocconi Students Fintech Society explores the use of machine learning (ML) to improve credit scoring accuracy. Traditionally reliant on statistical models, credit scoring is pivotal for banks to manage risk and determine loan terms. This study assesses several ML algorithms, including logistic regression, decision trees, random forest, and XGBoost, highlighting their effectiveness in differentiating between high-risk and low-risk borrowers using real data sets.

The integration of ML into credit scoring helps financial institutions enhance decision-making, reduce default rates, and comply with regulations against discriminatory practices. By automating and refining the evaluation of borrower reliability, ML provides a significant advantage in handling the increasing complexity and volume of financial assessments needed today.


 

Research Report

Read the Research Report below ⬇️



You can open the Python code file below through Google Collaborate ⬇️




 

Project Team

Project Leader: Gleb Legotkin

Analysts: Alexandra Minca, and Thomas Salvadego




Association Board :

Guillaume Abaz (President), Michelangelo Mauro (Vice President), Catharina Gärtner (Head of M&A), Gleb Legotkin (Head of Data Analysis), Matteo Nesiti ( Head of Operations).


Comments


bottom of page