Detection of Distributed Denial of Service Attacks Using Deep Learning Technologies

Abstract

Deep learning based ddos detection model Today, companies are facing with the huge network traffics mainly consisting of the various type of network attacks due to the increased usage of the botnet, the fuzzier, the shellcode or the network related vulnerabilities. This type of attacks are on the rise since they have negative effect on companies’ daily operations. By using the classification models, the attacks could be identify and separated earlier. The Distributed Denial of Service Attacks (DDoS) primarily focus on the preventing or reducing the availability of a service to innocent users. In this research, we focused primarily on the classification of network traffics based on the deep learning methods and technologies for network flow models. In order to increase the classification performance a model based on the deep neural networks has been used. The model used in this research for the classification of network traffics evaluated and the related metrics showing the classification performance have been depicted in the figures and tables. As the results indicate, the proposed model can perform well enough for detecting DDoS attacks through the deep learning technologies.

Publication
In The 5th High Performance Computing Conference.
Date