PERFORMANCE COMPARISON OF DDOS ATTACK MITIGATION TECHNIQUES USING AI ALGORITHMS

Abstract

In recent years, Distributed Denial of Service (DDoS) attacks have become a pervasive threat to network security, necessitating advanced mitigation techniques to safeguard online services. This study presents a comparative analysis of various DDoS attack mitigation techniques using artificial intelligence (AI) algorithms. We evaluated several models, including Decision Trees, Random Forests, Support Vector Machines (SVMs), Neural Networks, Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), and Deep Reinforcement Learning, based on their performance metrics: accuracy, precision, recall, and F1-score. Our results demonstrate that while traditional models such as Decision Trees and Random Forests offer solid performance, advanced AI models significantly outperform them. Neural Networks, CNNs, and RNNs exhibited superior capabilities in detecting and mitigating DDoS attacks, with Deep Reinforcement Learning achieving the highest overall performance. These findings underscore the efficacy of AI-driven approaches in enhancing DDoS attack mitigation strategies, providing valuable insights for future research and practical applications in cybersecurity.