Detection of Denial of Service Attacks against Domain Name System Using Machine Learning Classifiers

 Abstract—Domain Name System (DNS) provides name to address mapping services for the entire chain of Internet connectivity. Hackers exploit this fact to damage different parts of the Internet. In this paper we clarify possible Denial of Service (DoS) threats against DNS. An Intrusion Detection System (IDS) is introduced is the system to detect and classify different types of DoS attacks against DNS. This system consists of a statistical preprocessor and a machine learning (ML) engine. Three different types of neural network classifiers and support vector machines are evaluated in a simulated network. The results show that a backpropagation neural network engine outperforms other types of classifiers with 99% accuracy.