DDoS Attack Detection Using Ensemble Machine Learning Approach
Aditya Arya, Abhishek Kumar, Syed Safi Ahmad · 2023
As more devices connect to the internet, security becomes a major daily challenge. The prevalence of Distributed Denial of Service (DDoS) attacks, which are highly sophisticated, is a growing concern. These attacks are increasingly difficult to detect and are causing significant disruptions. They hinder legitimate users’ access to services, resulting in decreased network performance and, in severe cases, complete network shutdown. DDoS attacks reached a record high of nearly 13 million in 2022, while adversaries are increasingly skilled at bypassing traditional DDoS mitigation and defense mechanisms. In the last four years, over 4 million DDoS attacks have persisted for over an hour, with 25% of those lasting for more than 12 hours. This emphasizes the need for adaptive DDoS solutions that can effectively handle both short and long-lived attacks. This paper proposes an ensemble machine-learning technique which outperforms standalone machine-learning algorithms.