No 2 (2023)

The article deals with the infological aspects of the specialized hardware-software communication complex, which is part of the automated system of documentary exchange. The generalized architecture of the hardware of the complex and the main elements of the structure of its functional software are given. The structure of a number of software components is formalized in the notations of extended Petri nets. The formal definition of switching and the description of the organization of the transmission of documentary messages are stated.

A recommender intelligent system for the problem of health maintenance is presented, which provides the possibility of integral analysis of different types of information to assess the risk levels of chronic diseases and issue personal recommendations to reduce the likelihood or prevent the development of a pathological process. The system is based on a heterogeneous semantic network with a knowledge base solver that uses an argumentative reasoning algorithm. The recommendations given to the users are the practical realization of the modern concept of early detection of predisposition to the development of diseases with the focus on the active involvement of the person in the process of preventive measures. The results of testing the AI-HIPPOCRATES health-saving recommendation system on real data is considered; the possibility of using the proposed approach to assess the risks of chronic diseases (arterial hypertension, myocardial infarction, stroke, depression) based on a questionnaire is shown. At present, further research is aimed at modifying the developed intelligent recommendation system to monitor the psychoemotional state of critical infrastructure operators using remote monitoring data.

The article deals with the problem of ensuring the long-term preservation of legally significant electronic documents. A review of the problem and existing options for its solution is made, the problem of ensuring the long-term preservation of legally significant electronic documents is formalized as an optimal control problem under parametric disturbances. Document safety indicators and mathematical models for evaluating safety indicators are proposed. The result of the study are algorithms that allow, being implemented in the form of software and hardware, to control and organize long-term preservation. Examples of the practical implementation of the proposed approaches within the framework of large information systems are given. The advantage of the proposed algorithms and approaches is that their implementation will make it possible to comprehensively solve the important scientific and technical problem of organizing the long-term preservation of legally significant electronic documents. At the end of the article, further prospects for research aimed at solving the problem of long-term preservation of big data and systems built using distributed ledger technologies are identified.

The paper presents a method for local image contrast enhancement based on the distribution of gray levels in the vicinity of each individual pixel. The considered approach was automated using a genetic algorithm, which made it possible to eliminate the need for manual adjustment of the transformation parameters. The necessary criteria for assessing the quality of images are selected, among which the main ones are: the number of edge pixels, their total intensity, the measure of image entropy and the measure of brightness adaptation. Software components have been implemented and their functioning has been tested on various classes of images, which has shown the success of this approach for images with a high density of distribution of gradations of brightness, uniform illumination and a weak gradient of boundary pixels.

We study various models of measuring noises in the procedures of randomized entropy estimation of probability density functions: additive and multiplicative, measuring noises at the input and output of the object’s model. The properties of entropy-optimal probability density functions are studied, it is shown that the measurement noises corresponding to them are heteroscedastic.

The article discusses the features of solving problems related to the optimization of control processes in cyberphysical systems at manufacturing enterprises. With the help of the proposed approach based on the theory of graphs and structural matrices, there are opportunities to evaluate the effectiveness of the cyberphysical system and explore alternative options for its formation. The optimization task was formed in order to evaluate the effectiveness of management functions. It is considered as a linear programming problem with Boolean variables. The results show the possibility of improving the management efficiency of cyber-physical systems.

The paper derives the inference result for widely used fuzzy systems in case of non-singleton fuzzification. It is achieved by means of the approach based on the fuzzy truth values, which made it possible to reduce the computational complexity of the inference down to polynomial and to generalize the conditions for logical inference. The most commonly used defuzzification methods in applications were considered along with the obtained expressions of inference result.