+7 (495) 987 43 74 ext. 3304
Join us -              
Рус   |   Eng

Journal archive

№4(100) August 2022 year


IT management

Performance management

For a comprehensive assessment of the management decisions quality, it is necessary to take into account heterogeneous information presented both in numerical form and in natural language expressions. The effective occurs the use of data mining including neural network clustering and fuzzy set theory. The article presents our approach to the use of these methods for evaluating risks and the management decisions quality in Russian higher education on the example of the implementation of the most ambitious Project 5-100 for it. On the example, the expediency of the neural network clustering to assess the possibility of achieving the goals of any such large-scale project has been proved. Clustering the information database used for the analysis, makes it possible to carry out an objective selection of candidate universities-candidates for the right to receive state subsidies, as well as to adjust the composition of the Project participants. Another methods of intellectual analysis – the construction of a complex of fuzzy inference systems, – confirmed the possibility of a quantitative fi evaluating of the project based on the expert verbal estimates of the project. At the same time, the neural network clustering initially illustrated the unattainability of the Project 5-100 goals. The use of a complex of fuzzy inference systems confirmed this statement by the very low quantitative final assessment of the project on the basis of verbal expert opinions.

The research focuses on the development of applied software systems for automated environmental monitoring. The task of developing and integrating applied software, in particular calculation and analytical models based on machine learning (ML) methods, with an IoT platform of digital eco-monitoring for industrial enterprises is considered. Such a platform is used to create software and hardware systems of CEMS – Continuous Emissions Monitoring System class, designed for continuous monitoring of pollutant emissions into the atmospheric air at production facilities. Use of ML tools integrated with the platform allows to expand significantly the functionality of the existing CEMS, in particular to quickly build new SaaS services for forecasting the dynamics of pollution distribution. Given the high requirements for industrial systems, there is a need to create a specialized software product – an analytical server that implements the management of connected predictive analytical ML models with the required level of service quality, including automatic initialization of new analytical scripts as classes, isolation of individual components, automatic recovery after failures, data security and safety. The paper proposes a scheme of functional and algorithmic interaction between the IoT platform of digital eco- monitoring and the analytical server. The proposed implementation of the analytical server has a hierarchical structure, at the top of which is an application capable of accepting high-level REST requests to initialize calculations in real time. This approach minimizes the impact of one analytical script (class) on another, as well as extending the functionality of the platform in "hot" mode, that is, without stopping or reloading. Results demonstrating automatic initialization and connection of basic ML models for predicting pollutant concentrations are presented.

IT and education

Educational environment

The article provides a rationale for the relevance of developing a recommender system in the field of e-learning. The main approaches to building a recommender system are analyzed: collaborative, content and hybrid filtering. The main objects of the recommender system for choosing online courses are presented: the student, training modules (online courses), elements of knowledge that the user can receive at the end of the training. In algorithmic support, methods for creating recommender systems, such as machine learning, neural networks, genetic algorithms, are considered. Problems in the methods of building recommender systems have been identified: sparseness; cold start; scalability; searching for elements that are most likely to be preferred by the user from a common set of elements. The main problem of recommender systems is to obtain an accurate and high-quality recommendation for the selection of educational objects in accordance with user preferences. It is concluded that it is necessary to build an architecture of a recommender system, including a model of an individual learning trajectory. Filtration of educational objects occurs with the help of a genetic algorithm. The expediency of using a microservice approach to create a web application is determined. The functional tasks of the developed system are highlighted, such as data collection, analysis of user requests, the formation of educational objects using an individual learning trajectory and the issuance of recommendations for choosing online courses. An algorithm for the functioning of the recommender system, a scheme for the operation of the recommender system, as well as information support for the operation of this system have been developed. A general approach to the development of a universal recommender system that can be integrated into the client's service is proposed. The purpose of developing a recommender system for choosing online courses is to provide students with the most appropriate learning objects (sequence of objects) to study in accordance with the characteristics of the student and fragments of knowledge (competencies).

Software engineering

Algorithmic efficiency

Communication network simulators are software designed to model, explore, test and debug network technologies, including wireless decentralized self-organizing networks or ad-hoc networks. They greatly simplify the research, development and optimization of routing protocols in these networks. However, the well-known simulators have a number of disadvantages, including the difficulty of adding custom extensions to ad-hoc network routing protocols, the lack of the necessary network stack, the lack of routing algorithm visualization modes, low performance, and difficulty in debugging communication protocols. The purpose of this work is to create a simulation model of a wireless network that would allow us to explore, debug and evaluate the developed algorithms and routing protocols for ad-hoc networks. At the same time, the requirements for interface ergonomics and the ability to visualize the operation of algorithms, ensure the collection of statistics, and create various scenarios for the operation of the network come to the fore. The article proposes the structure of the simulation model, which includes the modules of the network subscriber, application software, network layer of the OSI data transmission model, radio module, radio transmission environment, statistics collection, visualization and scenario management. To solve the tasks set, the approach of discrete-event modeling was used. To create a simulator of wireless decentralized networks and routing algorithms, a set of classes was developed that implement the modules of the simulation model. Based on the proposed structure, module classes and discrete event simulation algorithm, a software implementation of the simulation model was created using the C++ programming language and the Qt framework. The developed simulation model was used in the course of an experimental study of the effectiveness of the network routing algorithm. The proposed software will simplify the development and debugging of algorithms and routing protocols for ad-hoc networks.

Currently, artificial intelligence is widely used in the formation of social, economic and environmental forecasts. When creating artificial intelligence, machine learning technologies, deep learning technology and searching for patterns in information arrays (Big Data), artificial language processing and generation technologies, etc. are widely used. At the same time, the issue of using artificial intelligence in scientific and technological forecasting has not been worked out enough. The purpose of the study was to find effective approaches to the use of artificial intelligence technologies in the formation of scientific and technological forecasts. The objective of the study was to identify artificial intelligence technologies that can be used at various stages of the life cycle of scientific and technological forecasting and to specify individual ways of using them to solve problems of predicting the level of development of science, engineering and technology compared to the world. This confirms the relevance of the study. The main research method is the analysis of domestic and foreign publications and best practices for using artificial intelligence technologies in scientific and technological forecasting, as well as the results of research work performed by the authors in the field of scientific and technological forecasting and adapting them to improve the formation of forecasts in the context of digital transformation of the economy and enterprises The authors considered the structure of artificial functions performed by technologies and identified priority areas for the use of artificial intelligence at various stages of scientific and technological forecasting. The expediency and features of the use of semantic analysis and cognitive technologies in predicting the level of readiness of equipment and technologies in comparison with the world under various scenario conditions are shown, which provides the greatest efficiency of the adopted solution. The issues of information and analytical support for the use of artificial intelligence in scientific and technological forecasting based on information technologies for decision support are considered. The novelty of the presented results lies in the fact that, for the first time, the authors describe the possibilities of using the most effective artificial intelligence technologies at various stages of the life cycle for the formation of scientific and technological forecasts from the standpoint of a systematic and integrated approach.

Models and methods

In the field of machine learning, there is no single methodology for data preprocessing, since all stages of this process are unique for a specific task. However, a specific data type is used in each direction. The research hypothesis assumes that it is possible to clearly structure the sequences and phases of data preparation for text recognition tasks. The article discusses the basic principles of data preprocessing and the allocation of successive stages as a specific technique for the task of recognizing ABC characters. ETL set images were selected as the source data. Preprocessing included the stages of working with images, at each of which changes were made to the source data. The first step was cropping, which allowed to get rid of unnecessary information in the image. Next, the approach of converting the image to the original aspect ratio was considered and the method of converting from shades of gray to black and white format was determined. At the next stage, the character lines were artificially expanded for better recognition of printed alphabets. At the last stage of data preprocessing, augmentation was performed, which made it possible to better recognize ABC characters regardless of their position in space. As a result, the general structure of the data preprocessing methodology for text recognition tasks was built.

The possibilities of a multi-agent approach for managing urban parking space are considered, which allows you to adequately represent the parking space and effectively solve the following tasks: monitoring congestion, searching and booking available parking spaces; building routes and navigation to selected places; parking; payment for parking services; monitoring compliance with parking rules; control and access control in closed parking lots (equipped with entrance and exit terminals and barriers); forecasting the main parameters, such as workload, income, turnover; informing users. The necessity of intellectualization of urban parking space management processes based on the use of methods and technologies of multi- agent systems (MAS), the main objectives of which are to: reduce the search time for parking spaces; increase the speed of traffic in paid parking areas; increase the turnover of parking spaces; reduce traffic congestion, fuel costs; reduce the number of parking violations on the road network; reducing the flow of personal vehicles entering the toll zone and stimulating the use of urban public transport; reducing environmental pollution. The greatest difficulty is the tasks of organizing the interaction of agents of various typologies in the collective solution of tasks, since each agent solving a specific task has only a partial idea of the overall task and must constantly interact with other agents. The features of prototyping MAS with an emphasis on modeling the interaction of certain types of intelligent agents in the problem area under study are presented. The obtained simulation results are the basis for the continuation and further development of research and development to create the final prototype of a MAS for urban parking space management.

Processes and systems modeling

The article examines the dependence between the Russian ruble exchange rate and oil prices with the use of neural network modeling. The relevance of the study can be confirmed by the interest of the monetary authorities in modeling the dynamics of the exchange rate for developing monetary policy measures. The research objective of the article is the estimation of the relationship between the Russian ruble exchange rate and oil prices using multilayer perceptron and recurrent neural network models. Moreover, the influence of additional factors, including foreign exchange interventions and geopolitical risks, is estimated. The results show that neural networks provide sufficient accuracy in estimation of the target variable. Furthermore, during the periods with foreign exchange interventions and high geopolitical instability there was confirmed a decoupling of the examined variables. The modeled time series preserve non-linear nature of exchange rate data generating process, as well as the asymmetry in the reaction of the ruble exchange rate to oil price shocks. The hyperparameters selection, use of bootstrap and ensembles of neural networks provide more robust estimates and confidence intervals for the oil price elasticity of the ruble exchange rate. Therefore, the combination of the aforementioned methods makes it possible to draw meaningful economic conclusions based on the trained neural networks, avoiding the problem of neural network weights non-interpretability.


Researching of processes and systems

Author: Andrey Shorikov

The solution of the problem of forecasting the state of complex socio-economic systems is possible only on the basis of appropriate dynamic economic and mathematical models that describe their main parameters, the presence of control actions and risks. In this paper, it is proposed to use a deterministic minimax approach for modeling and solving the problem of estimating the predicted states of a production system in the presence of risks. To make managerial decisions at a manufacturing enterprise aimed at improving the efficiency of its functioning, it is necessary to have high-quality information support, the basis of which is the solution of the corresponding problem of predicting the states of its basic parameters. In this article, to describe the functioning of a production system, it is proposed to use a discrete-time controlled dynamical system in the presence of risks. It is assumed that the values of the control action (admissible control scenarios) are realized from a finite set of admissible elements of the corresponding finite-dimensional vector space, and the realizations of the values of the phase vector of the model and the risk vector are limited by the given compact polyhedrons in the corresponding finite-dimensional vector spaces. Application of the developed discrete-time controlled dynamical model that describes the output products of an enterprise in the presence of risks, and the developed methodology for the formation and minimax estimation of the predictive set of its phase states in a given period of time, allow us to develop appropriate numerical algorithms that can be used in the development and creation of computer intelligent information systems that provide support for making effective management decisions at manufacturing enterprises. The main results of this work is the development of a new economic-mathematical model that describes the dynamics of the output products of an enterprise in the presence of risks and the creation on its basis of a methodology for constructing and minimax estimation of the predictive set of its phase states in the form of implementing a finite number of one-step operations that allow their algorithmization. The results obtained in this work can serve as a basis for developing methods for optimizing the management of enterprise production processes and creating computer intelligent information systems to support managerial decision-making.

Nowadays the introduction of robotic systems is one of the most common forms of the technological operations automation in various spheres of human activity. Among the robotic systems a special place is occupied by sequential multi-link robotic manipulators (SRM). SRM have become widespread due to relatively small dimensions and high maneuverability, which makes their use indispensable to solve various tasks. In practice, the effectiveness of the functioning of the SRM can be influenced by various types of external environment fuzzy factors. Among the external factors there is a group affecting the ability to determine the exact target position. Such factors often affect technical vision systems. This problem is especially relevant for special purpose mobile robots operating in aggressive environmental conditions. A situation similar to the described one also occurs when a medical robot manipulator is used for minimally invasive surgery, when the role of the control and monitoring system is assumed by an operator. In this regard, the organization of effective control taking into account influence of the external fuzzy factors, that prevent the correct recognition of the target position of the SRM instrument, is an urgent problem. The authors consider the solution of the inverse kinematics problem for SRM based on the use of fuzzy numerical methods, taking into account the possible occurrence of singular configurations in the process of solving.