Machine Learning in Banking: Different Approaches to Credit Risk

Analytical paper by MSc. André Monteiro (Banco do Brasil, Technology Directorate, Artificial Intelligence Sector) — a systematic review of machine learning applications in banking with emphasis on credit risk, covering 57 articles published between 1996 and 2019.

What you will find:

• Comparative analysis of ML models applied to credit risk: Logistic Regression, Artificial Neural Networks, Decision Trees, SVM, Discriminant Analysis and Genetic Algorithms
• A scoring grid ranking the best performing models — Logistic Regression confirmed as top performer (163 points), followed by ANN (159) and Decision Trees (100)
• Practical benefits for credit operations: monitoring of credit-risk variables, agility and uniformity in the credit process, improved default forecasting and better identification of money laundering and corruption crimes

Format: PDF, 8 pages, English. Immediate download after purchase.