** Cyber threats are evolving, and so should our defenses!** This project leverages the power of Ensemble Boosting Machine Learning to detect and mitigate DDoS attacks with 98.42% accuracy.
🔹 DDoS attacks are among the most disruptive cyber threats, crippling businesses and online services.
🔹 Traditional detection methods often fail against new & evolving attack vectors.
🔹 Solution? A robust AI-driven model that detects malicious traffic in real-time.
✅ Ensemble Boosting Model for high-accuracy classification
✅ Trained on real-world DDoS attack data (CICIDS2017)
✅ Identifies multiple attack types: TCP Flood, UDP Flood, HTTP Flood
✅ Achieves 99.99% Accuracy, 99.9% Precision, and 99.8% Recall
✅ Fast & Scalable Solution for real-time deployment
This system learns from network traffic patterns to distinguish between legitimate and malicious activity.
1 Feature Extraction - Identifying key traffic parameters (packet rate, byte rate, protocol distribution)
2 Model Training - Applying Boosting-based ensemble learning for better accuracy
3 Performance Evaluation - Measured using Accuracy, Precision, Recall, F1 Score & MCC
Base Model Used: Random Forest Classifier with boosting techniques
This model is trained on CICIDS2017, a well-known dataset in cybersecurity research. It contains real-world DDoS attack traffic, including:
🔹 TCP SYN Flood
🔹 UDP Flood
🔹 HTTP-based DDoS Attacks
Dataset Link: CICIDS2017 - Canadian Institute for Cybersecurity
🔹 Python 3.12
🔹 Pandas, NumPy
🔹 Scikit-learn
🔹 Matplotlib
Clone the repository:
git clone https://github.com/your-username/DDoS-Attack-Detection.git
cd DDoS-Attack-Detection
pip install -r requirements.txt python train_model.py python evaluate.py Connect with me!
📧 Srishti Sharma - [[email protected]]
🔗 LinkedIn - [link]
🔗 GitHub - [link]
Cyber threats aren’t stopping anytime soon—let’s stay ahead with AI-powered security!