A Study on the Application of Artificial Intelligence in DDoS Attack Defense: A Literature Review
Zhe Ping, Dingyang Jiao · 2024
As DDoS attack patterns become increasingly varied and challenging to detect with the advancement of the field, this paper aims to explore the possibilities of applying machine learning algorithms for the defense and detection of DDoS attacks. We selected papers published since 2000 concerning the aforementioned applications, investigated the algorithms described therein, and compared their data performance in the testing sections to explore the advantages and disadvantages. Our findings reveal that SVM exhibits excellent performance in detecting attack traffic, whereas decision trees and their enhanced versions are less prominent. Additionally, genetic algorithms are not suitable for attack traffic detection but excel in filtering irrelevant traffic features. These insights hold significant implications for both real-world applications in enhancing cybersecurity measures and theoretical contributions by laying groundwork for future research in the rapidly evolving domain of DDoS attack mitigation. In conclusion, this study offers a comprehensive overview of the landscape of machine learning in DDoS defense, providing valuable insights and directions for future investigations.