README.md

metadata

title: FraudGuard - Fraud Detection System
emoji: 🛡️
colorFrom: blue
colorTo: purple
sdk: gradio
sdk_version: 6.0.2
app_file: app.py
pinned: false

🛡️ FraudGuard - Intelligent Fraud Detection System

A machine learning-based fraud detection system with interactive Gradio interface.

🔍 Features

• Real-time Fraud Detection: Analyze transactions instantly
• Interactive Dashboard: Visualize risk distribution and patterns
• Batch Processing: Process multiple transactions at once
• Alert System: Get notified about high-risk transactions
• Detailed Reports: Generate performance analytics

🚀 How to Use

1. Single Transaction Analysis:

   • Enter transaction details in the "Process Transaction" tab
   • Click "Analyze Transaction" to get risk assessment

2. Quick Actions:

   • Use pre-defined buttons for common transaction types
   • Test with different risk levels

3. Batch Processing:

   • Select number of transactions
   • Click "Process Batch" for bulk analysis

4. View Dashboard:

   • Check real-time metrics and charts
   • Monitor recent transactions and alerts

🤖 Model Information

• Algorithm: Random Forest Classifier
• Accuracy: 85% (on training data)
• Features: 23 transaction and user features
• Training Data: UCI Credit Card Default Dataset

📊 Performance Metrics

• Real-time fraud probability scoring
• Risk level classification (Low/Medium/High)
• Processing time < 50ms per transaction
• Interactive visualizations with Plotly

🛠️ Technical Details

• Built with Gradio for interactive UI
• Scikit-learn for machine learning
• SQLite for data storage
• Plotly for visualizations
• Optimized for Hugging Face Spaces deployment

🔧 Development

To run locally:

pip install -r requirements.txt
python app.py