A Comparative Analysis of ML-Based Ensemble Techniques for DoS Attack Detection in IoT Environment

The fifth generation of technology is witnessing an increasing use of IoT (Internet of Things) devices, attributed to their ease and efficiency across numerous sectors, including healthcare and industry. However, they are also susceptible to cyber dangers such as Distributed Denial of Service (DoS) attacks. A DDoS attack is a significant threat to all layers of IoT infrastructure. Attack detection and mitigation by Machine Learning (ML) techniques are more accurate and efficient. However, the power of ML increases by ensemble methods. This study evaluates and compares ML based ensemble techniques such as bagging, boosting, and stacking for the detection of DDoS attacks. The efficiency of these techniques is evaluated using a benchmark dataset specifically designed for IoT Networks. This research clearly shows the benefits and limitations of each ensemble approach for IoT network-based DDoS attack. This research helps to ensure IoT security against emerging cyber threats.