The Use of Deep Neural Networks to Recognize Network Traffic Abnormalities in Enterprise Information and Telecommunication Systems
O.S. Amosov, Svetlana Gennadievna Amosova, Y. S. Ivanov, С.В. Жиганов · 2019 Twelfth International Conference "Management of large-scale system development" (MLSD) · 2019
It is proposed to recognize the network traffic abnormalities by using different deep neural network architectures. The experiment in detecting DoS attacks was conducted. The influences of neural networks layers combinations on the accuracy characteristics are revealed. New is the strengthening of the classification, through the reinforcement of an input vector to its cluster evaluation.