Deep Ensemble Learning With Pruning for DDoS Attack Detection in IoT Networks

The objective of this project is to develop an efficient and scalable DDoS attack detection system for IoT networks using deep ensemble learning methods. By integrating models such as Stacking Classifier (Random Forest, Gradient Boosting), Voting Classifier (Logistic Regression, Decision Tree, K-Nearest Neighbors), Naïve Bayes, Adaboost Classifier and the TPOT Classifier, the goal is to enhance detection accuracy while maintaining computational efficiency. Pruning techniques will be applied to reduce model complexity, ensuring the system is lightweight and suitable for resource-constrained IoT environments. The ultimate aim is to provide real-time detection of DDoS attacks with high accuracy and minimal overhead.