Shibaikin Sergei

RESEARCH, TEACHING, or OTHER INTERESTS

Computer Science, Artificial Intelligence

Scopus Publications

Application of machine learning methods for automated classification and routing in ITIL
VV Nikulin, S D Shibaikin, and A N Vishnyakov
IOP Publishing
Abstract The article analyzes the application of machine learning methods for automated classification and routing in ITIL library. ITSM technology and ITIL are considered. The definitions of the incident and IT services are given. Then, the vectorization and extraction of keywords in the information written in natural language is carried out and lemmatization and TF-IDF measure will be used. A comparative analysis of the application of machine learning methods is given as well as a comparison of the results of automatic classification of text information using gradient boosting and a convolutional neural network. Various parameters of these methods are considered and the most effective method of machine learning is determined. The results of using machine learning methods for automated classification of incidents allows high-precision routing of requests for restoring the operability of IT services, reducing response time and errors associated with the human factor.


Software library for the lighting fixtures optical characteristics calculation by the computer-aided design system kompas 3d
Sergey D. Shibaikin and Evgeny G. Alekseev
Redakcia Zhurnala Svetotekhnika LLC
The advantages and disadvantages of existing solutions for calculating lighting fixtures (LF) are discussed in the paper. The methods for solving the global illumination problem are demonstrated. The most important illumination models used to calculate LFs have been analyzed. The Torrance-Sparrow model is used as the primary model of reflective surface. The triangulation method based on approximation of the LF surface by a grid of triangles is used to increase the efficiency of LF design and computation. The optical ray propagation in a 3D optical system was modelled using classical laws of optics and the Monte Carlo method. The structure and the main steps of working with the library for CAD Kompas are described. The approximation of LF surface for different types of light sources and reflector parameters is presented. The implementation of approximations with subsequent ray tracing has demonstrated a good convergence of the problem solution. The results of the study have the potential to be useful for design engineers involved in the design and calculation of LFs, as well as developers of applied software in the field of computer graphics and computer-aided design systems.