Comparative Analysis of Feature Selection Techniques for Detecting Low-Rate DDoS Attacks in IoT Network
Dušan Drinić, Marija M. Novicic, Goran Kvaščev · 2025
With the rapid development of modern technologies and devices, there has been an expansion in the application of Internet of Things (IoT) networks across various domains. As the presence of these networks increases, so do the associated security risks. Given that the devices constituting these networks are vulnerable in terms of resource usage, a significant threat arises from attacks such as Low-Rate Distributed Denial of Service (LRDDoS), which often go undetected by traditional security systems. This paper presents a comparative analysis of various feature selection techniques for detecting LR-DDoS attacks within IoT networks, using the Extreme Gradient Boosting (XGBoost) algorithm. The results show no significant decrease in classification performance on reduced dataset compared to the full dataset.