Vol 14, No 1 (2024)

Terms, concepts and their content are not only the foundations of public communications, the keys to mutual understanding, knowledge transfer and cognition, but also fundamental elements in the construction of computer ontologies for artificial intelligence systems. The evolution of knowledge in subject areas leads, as a consequence, to the evolution of language, which is forced to change, borrowing words from other areas, expanding their meanings. The concept of systems analysis has a broad interpretation and is actively used in various fields of activity (science, economics, design, engineering, etc.), going deeper and adapting to specific subject and subject areas. The concepts related to system analysis (system, subject area, problem area, etc.) are also evolving. Systems analysis focuses on the relationships between elements of a system and determining its structure and functions; it includes various modeling and optimization methods and is used to solve specific problems and problems. Ontological analysis is associated with the study and description of objects, events, processes and relationships in the subject area, of which the system (or systems) may be a part; describes the structure and properties of the elements that make up the system and the relationships between them; used to create a formal description of a system and provide a general understanding of its components. The result of ontological analysis can be the basis for systems analysis or a part of it. Systems and ontological analyses are aimed at understand complex systems and their components and can be used together. The difference between these concepts is largely determined by the content of the field of study concepts, its boundaries, the object of research as a process, goal setting, the traditions and context that have developed among researchers in specific subject areas. Within the framework of the accepted nomenclature of scientific specialties of the Higher Attestation Commission, it can be argued that ontological analysis is included in the scientific specialty “Systems analysis, control and information processing, statistics.”

The article proposes an ontology of designing specialized digital twins of spatially distributed components of industrial-natural complexes (INC) based on the theory and intellectualized situational modeling system previously developed by the author. The ontology makes it possible to study the interaction of non-stationary INCs in normal operating modes and in the event of abnormal or emergency situations. The distinguishing feature of the accomplished development lies in the possibility of the preventive security analytics for the INC integration into existing infrastructures and provides a quantitative assessment of the effectiveness of proposed measures to prevent failures of INC components, including early detection of dependent (complex, cascading) failures. In order to increase the flexibility of modeling newly created INCs, the concept of structural security is presented, which generalizes ways to take into account various known aspects of security. The relevance of this task is determined by the growth in the number and power supply of INCs in the modern world, the complication of their interactions and the corresponding increase in the possibility of the most dangerous dependent failures, as well as the growth in the volume of data available for analysis as a result of the rapid development of the Internet of Things. The novelty of the proposed approach lies in the comprehensive application of expert knowledge at all stages of modeling in a cause-and-effect paradigm in order to synthesize preferred options for implementing INC structures.

Open communities of users in social media are a source of data that quickly presents the thematic agenda of issues relevant to the population. The indicators of user activity are views, likes, comments and reposts, and they are of a dynamic nature. The article presents a new vision at the topic modeling problems, the results of which are examined for dynamic properties. These data are relevant to solve problems of information support for regional and municipal development. The authors reveal their experience in designing an information system for integrated topic analysis of large open social media data. The system is based on three technologies: building dynamic topic models for monitoring social media, intelligent analysis of topic modeling results; and cognitive visualization of dynamic topic modeling results. To take into account design uncertainty, object modeling tools, system design and a modular approach were used.

The paper presents the results of experiments to assess the machine learning methods applicability for solving the problem of identifying argumentative connections in scientific communication texts. Argumentative connection is understood as a relationship that connects the premise and the conclusion of a typical reasoning or an argument used by the author to persuade the readers. To assess the quality, the characteristics of accuracy, completeness and F-measure were used obtained when solving the problem of recognizing argumentative connections between adjacent text fragments of two types: sentences and clauses. The basis of the experiment was a Russian-language corpus of texts from the field of scientific communication with arguments marked up by linguistic experts. For markup, the ArgNetBank Studio tool was used, which allows creating collections of texts with detailed argumentation markup. Data sets for machine learning were built on the basis of labeled texts, in which the ratio of pairs of text fragments (sentences or clauses) connected and non-connected by argumentative relationships was 1 to 3. To improve the quality of model training, the sets were balanced in two ways. In the first case, a balance was achieved due to the fact that an equal number of pairs of both types were selected from each text; in the second, pairs were duplicated. Using the obtained data sets, experiments were carried out on linking text fragments using different types of machine learning methods. The range of changes in quality assessments when recognizing related fragments depending on their share in the training and test collections was experimentally determined. It has been established that, within the framework of the existing imbalance in real collections, the values of quality assessments can vary within 40–50%. The novelty of the work lies in the study of the range of possible discrepancies in quality assessments when applying different machine learning methods on balanced and unbalanced training and test collections in Russian-language material.

The problem of ensuring the relevance of knowledge about the business process of an enterprise based on its ontological model is considered. The obsolescence of enterprise business process models leads to the need to redesign solutions for their automation. Approaches to solving the problem of maintaining the relevance of the business process model are presented. The regulatory basis of the business process of an enterprise was examined for its similarity to models developed to solve automation problems, and a conclusion was made that they have a common basis. This makes it possible to consider the regulatory basis as a metamodel and store data using linguistic variables. Ontological models built according to the IDEF5 standard for the most common regulatory documents of a business process are considered. It was been revealed that one of the most important elements of managing an industrial enterprise and its business processes is maintaining the relevance of all components, including the situational ontological model. A methodology is proposed for integrating an ontological model into the process of updating enterprise regulatory documents, preserving knowledge about the business process and retrieving it to generate changes to regulatory documents and models used to automate business processes. An example of using the technique for planning equipment repairs carried out outside the main work plan is considered.

To ensure the smooth operation of all existing electric power systems, it is necessary to periodically diagnose electrical equipment using various methods and models that take into account all possible parameters and factors affecting the condition. The paper considers an approach to assessing the technical condition of electrical equipment using weighted fuzzy rules, taking into account different types of information (measuring, expert). The novelty of the approach is the representation of parameters in the form of fuzzy triangular numbers and the formation of importance weight vectors of term sets parameter values. This makes it possible to accurately assess the technical condition of electrical equipment in the context of different types of information to predict the condition of the equipment; to quickly identify parameters which values are outside acceptable limits, thereby determining the preliminary cause of equipment failure, as well as making informed diagnostic decisions regarding the condition of electrical equipment.