Section
MATHEMATICAL MODELING, NUMERICAL METHODS AND SOFTWARE SYSTEMS
V. N. Belovodskiy ON INTERPOLATION AND APPROXIMATION OF FUNCTIONS USING NEURAL NETWORKS.
UDC 519.6 DOI 10.24412/2413-7383-2024-3-4-19 Language: Russian Annotation: The evolution of neural networks and the ever-expanding field of their application dictate the need to study its theory basics at early stages of student learning. And it is natural to begin this process with consideration of typical problems of the approximation theory of functions in course of computational methods. Such tasks are discussed in this article. Using the example of a two-layer neural network, the possibilities of neural networks for interpolation and approximation of functions of one variable are studied. Computational experiments are being performed with the help of a specially developed program in which the error minimization of the neural network is carried out using a subroutine implementing Levenberg – Marquardt methods. Minimizing errors in the identification of network parameters is carried out by varying their initial values using the Sobol sequence. The results obtained are compared with classical algebraic approaches. The advantages and disadvantages of the neural approach are noted, the generalizations are made.
Keywords: interpolation, approximation, neural network, minimization, Sobol sequence.
V. S. Solod PHYSICAL AND MATHEMATICAL MODEL FOR THE DEVELOPMENT OF AN EXPERT SYSTEM FOR SECTION ROLLING PRODUCTION.
UDC 528.013, 621.771.07 DOI 10.24412/2413-7383-2024-3-20-28 Language: Russian Annotation: Based on original mathematical models of metal forming, flow stress, temperature distribution during hot deformation, heating and cooling of rolled products, the software package allows using previously developed technology options in new technological developments, which significantly reduces design time. In addition, this model can be used as the basis for an expert system for optimizing section rolling technology based on artificial intelligence.
Keywords: model of a section mill, software package, rolling technology, expert system.
Section
ROBOTS, MECHATRONICS AND ROBOTIC SYSTEMS
V. M. Zuev COMPARISON OF OBJECT DETECTION BY ARTIFICIAL INTELLIGENCE IN COMPARISON WITH CLASSICAL METHODS.
UDC 528.013 DOI 10.24412/2413-7383-2024-3-30-35 Language: English Annotation:
Nowadays, neural networks are increasingly being used to detect objects in a video image. The article presents a comparison of object detection methods using artificial intelligence and classical methods. The use of a neural network is not always justified. Traditional filtering methods often restore the image better at low signal-to-noise ratios, require less memory (which is important when implemented on microprocessors), have greater speed and load the processor less.
Keywords: artificial intelligence, filtering, neural network.
A. E. Pokintelitsa SUBSTANTIAL FOUNDATIONS OF THE MATHEMATICAL MODEL OF A DIGITAL HALFTONE IMAGE.
UDC 51-74:004.932 DOI 10.24412/2413-7383-2024-3-36-43 Language: Russian Annotation:
This paper analyzes the substantial foundations of the mathematical model of a digital halftone image. The process of the model construction is considered. It is indicated that the representation of visual information using a specially ordered set of values of a discretized image function which determines the brightness of the image makes it possible to use mathematical methods of image processing. The ordering function establishes the relationship between the individual elements of a digital image (pixels). It is concluded that the ordering function is the key to extracting useful information from a set of visual data that is being processed.
Keywords:digital image, mathematical model, image processing, visual information representation, modeling.
Section
SYSTEM ANALYSIS, CONTROL AND INFORMATION PROCESSING, STATISTICS
S. A. Bolshakova THE SYSTEM OF AUTOMATIC ADAPTATION OF RUSSIAN-LANGUAGE TEXTS AND ITS PRACTICAL SIGNIFICANCE.
UDC 004.912 DOI 10.24412/2413-7383-2024-3-45-54 Language: Russian Annotation:
The article describes an adaptation system using a method of lexical simplification of the Russian-language text using a specially marked database of synonyms and a set of rules that allow automatic lexical replacement of words and phrases while preserving the semantics of the text and restoring the correct syntax. Dictionaries that are in the public domain were used to form a marked-up database of synonyms. The proposed methods and algorithms, as well as a database of synonyms, can be used for adaptation tasks, search engine optimization and automatic text abstraction, as well as for automatic translation.
Keywords: natural language processing, automatic text analysis, synonymy, automatic adaptation.
B. V. Pavlenko, V. I. Bondarenko AN INTELLIGENT ALGORITHMIC METHOD FOR CALIBRATING SIGHTS.
UDC 004.82+004.891 DOI 10.24412/2413-7383-2024-2-59-77 Language: Russian Annotation:
The article presents the development process and the results of the semi-autonomous experimental intelligent algorithmic optical sight calibration system using the mildot grid. The features of creating your own datasets for specific areas are touched upon. The recognition system is based on the use of fast neural networks such as YOLO.
Keywords: intelligent algorithm, calibration, optical sight, YOLO.
V. Yu. Ustenko OPTIMIZATION OF PRELIMINARY COMPETENCE ASSESSMENT OF CANDIDATES IN IT TECHNOLOGIES USING MICROSOFT ML BUILDER.
UDC 528.013 DOI 10.24412/2413-7383-2024-3-64-69 Language: Russian Annotation:
The article explores the potential use of Microsoft ML Builder to optimize the preliminary assessment of candidates' competencies in the IT field based on their academic achievements. Leveraging machine learning enables HR departments to automate the hiring process, enhancing its objectivity and efficiency.
Keywords: machine learning, personnel selection, HR, academic achievements, Microsoft ML Builder.
R.S. Khakimov, O.L. Nizhnikova, M.V. Blizno ON THE ISSUE OF DEVELOPING A DATA ANNOTATION SYSTEM FOR COMPUTER VISION TASKS.
UDC 004.93/004.932 DOI 10.24412/2413-7383-2024-3-70-79 Language: Russian Annotation:
The article provides an analytical overview of software solutions for annotating data in order to highlight their functionality, identify advantages and disadvantages that should be taken into account when creating your own annotator.
Keywords: data annotation system, annotator, data markup.
R. S. Khakimov, Y. S. Pikaliov DEVELOPMENT OF METHODOLOGY FOR SELECTION OF GRAPHICAL SHELL IN PYTHON PROGRAMMING LANGUAGE BASED ON HIERARCHICAL ANALYSIS METHOD.
UDC 004.4:519.816 DOI 10.24412/2413-7383-2024-3-80-86 Language: Russian Annotation:
The article discusses the key points and features of the hierarchy analysis method, the main stages of the method application: formation of the hierarchical structure of the task, pairwise comparisons of criteria and alternatives, checking the consistency of evaluations and calculation of global priorities. Functionality, cross-platform, resource intensity and widget design are identified as criteria for evaluating libraries. The application for selecting a suitable Python library for creating a graphical user interface is justified.
Keywords: hierarchy analysis method, criteria, alternatives, library.
Section
INTERDISCIPLINARY AI RESEARCH. COGNITIVE ASPECT
K. S. Ivashko, V. N. Piguz, S. A. Izosimova THE COGNITIVE ASPECT OF WARTIME STRESS RESEARCH.
UDC 004 DOI 10.24412/2413-7383-2024-3-86-101 Language: Russian Annotation:
The article discusses new models of psychological assistance that take into account the specifics of wartime and allow to provide assistance both individually and in groups. It is important to note that the development of theoretical foundations and computerized intellectual technologies of diagnostics and self-regulation of psychoemotional states. Conditions of chronic wartime stress is a complex approach, requiring a deep understanding of both psychological and technical aspects. New models of psychological care must be adapted not only to the conditions of wartime conditions, but also to the different cultural, social, and individual peculiarities of each person. Therefore, it is important to create innovative approaches to help psychologically affected people, taking into account the specificity of the time, in which they are.
Keywords: stress; self-regulation; technology; state; development.