Fraud_Detection_System

A machine learning based fraud detection system that identifies suspicious financial transactions using Random Forest and Isolation Forest algorithms.

🚀 Features

• Detects fraudulent transactions in real-time
• Uses both supervised and unsupervised ML models
• Analyzes transaction amount, location, time, and merchant behavior
• Classifies transactions as SAFE, FLAGGED, or BLOCKED
• Generates detailed risk factor reports

🛠️ Tech Stack

• Language: Python
• Libraries: Scikit-learn, Pandas, NumPy
• Algorithms: Random Forest Classifier, Isolation Forest

⚙️ How to Run

1. Clone the repository git clone https://github.com/TVARDHINI/Fraud_Detection_System.git

2. Install dependencies pip install -r requirements.txt

3. Run the system python fraud_detection.py

📊 How It Works

• Random Forest → Supervised classification to detect known fraud patterns
• Isolation Forest → Unsupervised anomaly detection for unusual transactions
• Feature Engineering → Extracts time, location, amount, and merchant-based features