Explainable AI for Financial Fraud Detection and Credit Scoring

Project Overview

This project implements machine learning models for credit scoring and financial fraud detection and applies Explainable AI (XAI) techniques to interpret model predictions and improve transparency in financial decision-making.

Key Features

• Credit risk prediction using ML models
• Financial fraud detection
• Model interpretability using SHAP and LIME
• Transparent and trustworthy AI decisions

Technologies Used

• Python
• Jupyter Notebook
• Scikit-learn
• SHAP
• LIME
• Pandas, NumPy, Matplotlib

Dataset

German Credit dataset in ARFF format.

Results

The project demonstrates how explainability techniques help understand feature contributions and model behavior in financial ML systems.

How to Run

1. Clone the repository
2. Install required dependencies
3. Open the Jupyter Notebook
4. Run all cells sequentially