Estimation of applicability of modern neural network methods for preventing cyberthreats to self-organizing network infrastructures of digital economy platforms

• Published in: SHS Web of Conferences Vol. 44; p. 44
• Main Authors: Kalinin, Maxim, Krundyshev, Vasiliy, Zubkov, Evgeny
• Format: Journal Article Conference Proceeding
• Language: English
• Published: Les Ulis EDP Sciences 2018
• Subjects: Infrastructure; Neural networks
• ISSN: 2261-2424; 2416-5182; 2261-2424
• DOI: 10.1051/shsconf/20184400044

The problems of applying neural network methods for solving problems of preventing cyberthreats to flexible self-organizing network infrastructures of digital economy platforms: vehicle adhoc networks, wireless sensor networks, industrial IoT, “smart buildings” and “smart cities” are considered. The applicability of the classic perceptron neural network, recurrent, deep, LSTM neural networks and neural networks ensembles in the restricting conditions of fast training and big data processing are estimated. The use of neural networks with a complex architecture– recurrent and LSTM neural networks – is experimentally justified for building a system of intrusion detection for self-organizing network infrastructures.