DDoS Detection using Multilayer Perceptron

Abstract

In recent years, distributed denial of service (DDoS) attacks have grown to be a serious threat to network security, severely disrupting internet services and enterprises. Due to the dynamic and evolving nature of these attacks, detecting and mitigating them has become a difficult task. By examining the network traffic data, machine learning algorithms like Multilayer Perceptrons (MLPs) have demonstrated the potential in identifying DDoS attacks. This research study investigates the application of MLPs for DDoS detection and assess the model’s performance on a real-world dataset. We also examine how various hyperparameters affect the model’s performance and suggest an optimization technique to increase its accuracy. The outcomes of our research show that MLPs have the potential to be an effective tool for detecting and countering DDoS attacks, in addition to offering suggestions for future network security research