ISSN 0021-3454 (print version)
ISSN 2500-0381 (online version)
Summaries of the issue


An approach is proposed that allows to use the multi-model logical-dynamic description for solving simultaneously the problems of synthesizing the technology of control over cyber-physical systems and planning execution of relevant operations. An example of the problems in question related to organization of the work of industrial robotic complexes is considered; solutions using the developed software are presented
A poly-model description is proposed for functioning of cyber-physical systems. The systems under consideration are multifunctional hardware-software complexes designed to receive (transmit), store, and process information, as well as to generate control actions on the objects of service that perform a given set of targets and are not parts of the cyber-physical system. The description is based on a novel dynamic interpretation of the relevant processes. The stage of planning of measuring and computing operations carried out by the cyber-physical system to determine its location and position, is considered in view of special importance of the operations in solution to the problems of control over the objects. An information on the software complex providing solutions to the problem of operational planning of measuring and computing tasks, is presented
Methods of knowledge representation in the precedent knowledge base, as well as search and extraction of precedents similar to a current situation, are considered as applied to solving the problems of proactive management in decision-support systems. It is shown that for decision-support systems with large precedent bases, the most rational option is to combine the precedents extracting approaches based on the "k nearest neighbors" method. In this case, other methods are used for pre-filtering of pre-cedents in the database


The problem of automation of information-analytical support of the life cycle of organizations and their products is considered. As a result of the study of requirements and methods for solving the problem, it is proposed to use combined models and methods of the theory of artificial intelligence as the basis for computer-aided design of applied software that solves this problem
A model-oriented approach to the design of user interfaces for expert systems is proposed. Several provisions are given to substantiate the necessity of using declarative-procedural languages to describe user interfaces using a model-oriented approach to system design. A conceptual model of user interface for expert system is developed to describe the basic properties and components of the interface. A concept of dynamic reconfiguration of the user interface for the state of the computing process is presented. The proposed approach is considered as an initial step in formation of a new methodology for designing user interfaces using languages with visual notation
A classification scheme for image segmentation algorithms is considered, requirements for new algorithms are formulated. A three-stage block scheme is proposed for designing new algorithms. An algorithm of quasi-optimal segmentation is discussed as a version of the block scheme implementation. An attribute reflecting the features of the group of hierarchical algorithms that the proposed algorithm belongs to is introduced. Application of the introduced attribute is shown to refine the classification scheme and the requirements for the algorithms being created
Distributed systems of complex modeling using the Earth remote sensing data are considered. The basic principles of software designed to support automated creation and operation of such systems are presented, components of the complex modeling system are analyzed. The software feasibility is illustrated by the example of presented results of the developed system approbation in monitoring and forecasting of the Northern Dvina river flood


The field of event correlation for systems of security information and event management systems is investigated. The purpose of the research is to determine the types of information objects by analyzing the security event log of the infrastructure under study. A correlation approach based on definition of relationships between equivalent events properties by their mutual utilizing is proposed. The study of revealed relationships results in definition of the analyzed infrastructure in the form of types of high-level objects. Results of an experiment on the structural analysis of the Windows security events log are presented. The cases of unstable work of the proposed approach and their possible causes are described. The evaluation and interpretation of the obtained results testifying to the possibility of application of the presented approach in practice are given
A method of information processing on the base of Big Data technologies aimed at computer at-tacks detection is studied. The need to create specialized approaches and design methods that will im-prove the efficiency of processing the received information is justified. The possibilities and effectiveness assessments of parallel data processing with the purpose of computer influences detection using a functional approach, as well as the key principles of working with Big Data, are considered. The mathematical model by means of which the technique of intrusion detection is developed is presented. The principle of implementation of the tasks of information processing and anomaly detection based on integration of Hadoop, Snort platforms is described. Main results of the experimental evaluation of the method used to detect computer attacks are presented