ABSTRACTS OF ARTICLES OF THE JOURNAL "INFORMATION TECHNOLOGIES".
No. 7. Vol. 25. 2019

DOI: 10.17587/it.25.397-404

V. V. Bova, Associate Professor, e-mail: vvbova@sfedu.ru, Yu. A. Kravchenko, Associate Professor, e-mail: yakravchenko@sfedu.ru, E. V. Kuliev, Associate Professor, e-mail: ekuliev@sfedu.ru, V. V. Kureichik, Head of CAD department, e-mail: vkur@sfedu.ru, Southern Federal University

Behavior Modeling of the Subject in the Internet Services Based on the Bacterial Optimization Modified Algorithm

This article is devoted to the solution of the fundamental and interdisciplinary scientific problem of identifying information events that pose a threat in the process of the subject's interaction with Internet services and develop effective strategies for preventing or reducing undesirable consequences. A special place in the solution of this problem takes the process of building models of intelligent agents' behavior. The large dimension of the problem being solved and the high degree of information uncertainty do not allow the use of mathematical analysis methods. In this paper, a modified bacterial algorithm is used to obtain quasioptimal solutions. In contrast to the canonical form, the proposed algorithm introduces new mechanisms to support search procedures. The specific scientific result was the swarm model of agents' behavior whose purpose of search procedures is to assess in real time the correspondence of the events' attributes to possible negative precedents organized in the form of ontological structures. The obtained results are relevant and in demand when creating methods and algorithms for using artificial intelligence to simulate, predict and support effective and psychologically safe informational and educational activities of a person in the Internet space. The work is directed to the study of this problem and the development of intelligent assistant systems that optimize educational activities in the Internet space, based on the methodology of biologically plausible machine learning methods. The developed models of information and educational activities and scenarios of the subject's behavior, taking into account its individual characteristics, will form the basis for the development of intelligent assistant systems and automated tutoring systems. Biologically plausible machine learning methods that provide the most effective solution to the problems of classification, clustering, recognition, optimization and coding of information will make it possible to find solutions that are close to optimal for an acceptable time in conditions of various kinds' Internet resources avalanche-like growth, on the one hand, it has a purposeful, self-organizing and self-developing character, on the other — signs of uncertainty and risk.
Keywords: information threats, ontological structures, intelligent agents, swarm algorithms, behavior patterns

P. 397–404

To the contents