Development of a new system to detect denial of service attack using machine learning classification

Mohammad M. Rasheed (University of Information Technology and Communications)
Alaa K. Faieq (Baghdad College of Economic Sciences)
Ahmed A. Hashim (University of Information Technology and Communications)

Indonesian Journal of Electrical Engineering and Computer Science
Vol 23, No 2: August 2021

Denial of service (DoS) attack is among the most significant types of attacks in cyber security. The objective of this research is to introduce a new algorithm to distinguish normal service requests from the denial of service attacks. Our proposed approach can detect the denial of service attacks by the analysis of the packets sent from the client to the server, which depend on machine learning. Our algorithm collects different datasets of benign network traffic and different types of denial of service attacks, such as DDoS, DoS Hulk, DoS GoldenEye, DoS Slowhttptest and DoS Slowloris, that were used for training. Moreover, our algorithm monitors the network every specific time to find denial of service attack. Our results show that the algorithm can detect the benign cases and distinguish the types of denial of service attack. Furthermore, the results could achieve 99 percentage of correct classification of all selected cases.