A High-Precision Machine Learning Algorithm for Detecting DDoS Attacks in a Cloud Computing Environment
A Distributed Denial of Service (DDoS) attack is an evil attempt to flood a website or a network with malicious traffic to force it to slow down or stop working and it is now extending to other technologies like cloud computing, IoT, and edge computing. This attack has different types, the attacker can exploit the UDP protocol to flood the victim's devices with a huge amount of data, or exploit vulnerabilities in network protocols, or may target the application layer. All of the above attempts to overwhelm all available resources including memory, CPU, and potentially the entire network aiming to incapacitate the victim's machine or server. Despite numerous proposed defensive mechanisms, these mechanisms often fall short as attackers continuously adapt using new automated tools. This is why we propose a machine learning-based approach for DDoS attack detection in cloud computing environments. Using machine learning classifiers, Random Forest (RF) and K-Nearest Neighbors (KNN) and compared based on classification performance and computational efficiency. Experimental results showed that the Random Forest classifier achieved the best performance by reaching an accuracy of 99.97% with minimal false positives. Finally, integrate the best selected model into a Flask-based real-time detection system able to classify generated traffic as either Normal or DDoS Attack.