ABSTRACTS OF ARTICLES OF THE JOURNAL "INFORMATION TECHNOLOGIES".
No. 12. Vol. 28. 2022

DOI: 10.17587/it.28.644-653

S. M. Avdoshin¹, Cand. of Tech. Sc., Professor, D. V. Pantiukhin^1,2, Senior Lecturer, I. M. Voronkov^1,3,4, Invited Lecturer, Senior Research Fellow, Deputy Head of the Center of Neural Network Technologies, A. N. Nazarov¹, Dr. Tech. Sc., Professor, V. I. Muhamadiev³, Engineer, M. K. Gordenko¹, Leading Expert, Nhich Van Dam¹, Invited Lecturer, Ngoc Diep Nguyen⁵, Ph.D., Assistant of the Department of Mechanics and Control Processes,

¹ HSE University — National Research University Higher School of Economics,
² MIREA — Russian Technological University,
³ Centre of Information Technologies and Systems of Executive Authorities,
⁴ International Center of Informatics and Electronics,
⁵ RUDN Engineering Academy, Moscow, Russian Federation

Analysis of Neural Network Intrusion Detection Methods and Datasets for their Training

Approaches based on neural network classifiers to the detection of computer attacks are considered. The problems of training such classifiers are discussed. Data sets on computer attacks for wired and wireless systems are considered. The results of evaluating such sets by the degree of imbalance are given. The problems of learning on unbalanced data sets and approaches to balancing the training set in the case of rare attacks, including those using generative adversarial networks, are described.
Keywords: intrusion detection, neural network, training sample balancing, intrusion detection datasets

Acknowlegements: The work was supported by the RFBR grant, project No. 21-57-54002.

P. 644–653

To the contents