A Comprehensive Survey on the Efficacy of Machine Learning Algorithms for Detecting DDoS Attacks

Denial-of-service (DoS) and distributed denial-of-service (DDoS) attacks are blatant attempts to prevent legitimate users from reaching a service or from using the service they already are using. DDoS attacks are possible due to the internet's current design, and an attacker gain control over many machines by exploiting their weak factors and setting up different targeted networks, also known as “Botnets”. Once an attacker is ready with the targeted network, they can launch a massive, coordinated attack against a target or targets. As new attacks are constantly being created, and the number of hosts that may be attacked online continues to grow, several various DDoS attack detection and prevention methods have been developed to combat this growing threat. In this paper, we discuss a wide range of DDoS attacks and methods, as well as countermeasures. This review is useful because it addresses many different approaches to stopping DDoS attacks, such as detecting them, defending against them, minimizing their impacts, tracing their origins, answering unresolved problems, and overcoming research difficulties.