Vol 31, No 8 (2025)

The article discusses the issues of developing a decision support system in the field of regional development management. The aim of the study is to develop an approach to strategic planning of the socio-economic development of the region with the reconciliation of modern decision-making support tools. The objectives of the research are to develop the concept of this approach, to form an iterative algorithm for correcting the values of target indicators and an algorithm for intellectual decision support. It is also required to describe the functioning of the proposed intelligent adaptive simulation model and to test it. As a result of the research, the key elements of an intelligent decision support system are presented, algorithms for determining control parameters are developed. A model has been developed that includes a number of important elements, among which a special place is occupied by blocks responsible for developing decision parameters. In particular, at the level of formation of the management system, an algorithm for correcting the values of target indicators is proposed based on an analysis of the degree of implementation of the set plan and an assessment of the availability of an unused resource, characterized by the presence of iterative mechanisms and allowing to determine the parameters for adjusting target indicators of regional development. In turn, for the formation of management decisions, a two-stage algorithm of intellectual decision support has been developed, based on the identification of typical situations, characterized by the use of fuzzy boundaries of affiliation and allowing to determine solutions appropriate to the situation. The approbation was carried out on data from the regions of the Volga Federal District. A quantitative forecast of changes in the values of parameters for three scenarios up to 2030 was obtained. The average accuracy of forecasts that were obtained using the proposed model of the region turned out to be higher than a similar forecast without an intelligent adaptive simulation model by 2.8 percentage points. The resulting model can be used as part of the justification of government regulation measures.

The paper suggests a new interpretation of the three versions of the theory of SK-languages (standard knowledge languages) developed by the author of the present paper in several previous publications. The theory of SK-languages (SKL-theory) is the central constituent of the theory of K-representations (knowledge representations), or TKR, introduced in early 2000s by the author of the paper. The three versions of the SKL-theory are considered as mathematical models of the variety of semantic representations (SRs) of arbitrarily complex sentences and discourses in natural language (NL). The possibilities of interactions of several original expressive mechanisms of the SKL- theory while constructing SRs of the scientific notions’ definitions are demonstrated. The broadest prospects of describing semantic structure of sentences and discourses in NL pertaining to biology and medicine are indicated. A special attention is paid to explicating the advantages of TKR-based approach to building SRs of scientific definitions in comparison with Universal Conceptual Cognitive Annotation, Abstract Meaning Representation, and Uniform Meaning Representation. The second objective of the paper is to demonstrate the possibilities of an original algorithm of semantic parsing of the scientific notions’ definitions suggested in one of the previous publications of the author. The output of the algorithm are SRs of the scientific notions’ definitions being the expressions of SK-languages.

The problem of automatic recognition of gesture images for computer vision systems is considered. The process of preparing the initial data, creating a training and test dataset is described. A custom dataset has been created, as well as integration and preparation of external data. Research of popular neural network methods of sign language recognition has been conducted and an assessment of their effectiveness in the presence of negative factors has been obtained. Recommendations are given to improve the quality of gesture recognition.

Modern local area networks of the Wi-Fi standard have many advantages: high communication speed, mobility and a large number of simultaneously connected devices. Nevertheless, there is a significant drawback of this standard related to the small communication radius, which significantly limits the scope of this technology. First of all, this concerns the connection of Internet of Things (IoT) devices. For these purposes, it is necessary to develop separate standards, for example: LPWAN, NB-IoT, LoRA and others. This article substantiates the possibility of using the DECT standard as a universal standard for the IoT, digital telephony, etc. The paper formulates the basic requirements for a wireless network for local needs, such as an extended communication radius, an unfilled frequency spectrum, and energy-saving capabilities. It is shown that DECT satisfies all conditions well. The fundamental possibility of encapsulating MQTT messages in SIP messages is demonstrated. A new communication scheme has been proposed that allows for the transmission of MQTT control messages to an IoT sensor. It describes how the delivery of IoT control messages based on subscriber numbers in SIP can be implemented. The proposed communication scheme allows simultaneous provision of two communication services: radiotelephony and last mile for wireless connection of IoT devices. The application of the proposed solution for the creation of remote laboratories for the research and study of IoT networks and DECT technology is described.