Fraud Detection System

A machine learning project that detects fraudulent credit card transactions using a Random Forest Classifier. This system generates synthetic transaction data, preprocesses it, trains a model, and visualizes the performance using Confusion Matrices and ROC Curves.

📌 Project Overview

This project implements a complete pipeline for fraud detection:

1. Data Generation: Creates a synthetic dataset with a realistic class imbalance (approx. 5% fraud).
2. Preprocessing: Scales features using StandardScaler and handles train-test splitting.
3. Modeling: Trains a Random Forest Classifier (with class weighting to handle imbalance).
4. Evaluation: Outputs classification reports, accuracy scores, and visualizes results.

🚀 Key Features

• Synthetic Data Engine: Generates realistic transaction data including features like amount, time, and anonymized vectors (v1-v4).
• Imbalance Handling: Uses class_weight='balanced' to ensure the model learns to identify the minority class (fraud) effectively.
• Visualization: Automatically generates and saves a fraud_detection_results.png showing the Confusion Matrix and ROC Curve.
• Real-time Prediction: Includes a method to predict individual transactions.

🛠️ Technologies Used

• Python 3.x
• Scikit-Learn: For model building, preprocessing, and metrics.
• Pandas & NumPy: For data manipulation and synthetic data generation.
• Matplotlib & Seaborn: For plotting and visualization.

📦 Installation

To run this project, simply install the required dependencies:

pip install -r requirements.txt



💻 Usage
Run the main script to generate data, train the model, and view results:

```bash
python fraud_detector.py


Output:
    The script will output model performance metrics in the console and save a visualization image:

    Console: Prints the Confusion Matrix, Classification Report, and ROC-AUC Score.

    Image: Saves fraud_detection_results.png (Confusion Matrix + ROC Curve).

📊 Results:
     The model achieves high accuracy on the synthetic dataset. Below is an example of the performance metrics visualized:

     ROC-AUC Score: ~1.00 (Excellent separation between classes)

     Precision/Recall: High detection rate for fraudulent transactions with minimal false positives.

📝 License
This project is open-source and available under the MIT License.