Journal of Siberian Federal University. Humanities & Social Sciences / About the Possibility of Managing the Training Process Using Predictive Models Based on Artificial Intelligence

Full text (.pdf)
Issue
Journal of Siberian Federal University. Humanities & Social Sciences. 2025 18 (2)
Authors
Bazarin, Kirill P.; Bartsev, Sergey I.; Kovalev, Viktor N.
Contact information
Bazarin, Kirill P. : KGKU “Krasnoyarsk Institute for the Development of Physical Culture and Sports”; FSBSI FRC KSC SB RAS Separate subdivision “Institute of Biophysics of the Siberian Branch of the Russian Academy of Sciences” Krasnoyarsk, Russian Federation; ; Bartsev, Sergey I. : FSBSI FRC KSC SB RAS Separate subdivision “Institute of Biophysics of the Siberian Branch of the Russian Academy of Sciences” Krasnoyarsk, Russian Federation; Kovalev, Viktor N. : Siberian Federal University Krasnoyarsk, Russian Federation
Keywords
physical activity; sports; neural networks; artificial intelligence; digital twin; individualization of training; digitalization
Abstract

To assess the possibility of effective application of predictive systems based on artificial intelligence for planning the training process. The study involved 155 athletes, representatives of various sports. Male sex – 96 people, average age 24.34 ± 3.54 years; female sex – 59 people, the average age was 23.12 ± 2.3 years. The control group consisted of 101 people who did not experience high systematic physical exertion. Male 53 people, mean age 23.17 ± 2.54, female – 48. The average age was 22.12 ± 3.01 years. A complex of neural networks was formed, allowing to predict a number of key indicators of physiological reactions of the blood system in the dynamics of the annual training-competitive cycle in qualified athletes. A series of virtual experiments was carried out in which the possibility of avoiding the development of decompensation was studied by varying the acting factors. The average accuracy of the neural network model was 96.9 %, which is a fairly high indicator for predicting biological processes. The results of virtual experiments demonstrate a reliable correspondence to the subgroups in which the real acting factors corresponded to the model ones. The possibility of effective application of predictive systems based on artificial intelligence for planning the training process has been demonstrated

Pages
410–416
EDN
EPZXKE
Paper at repository of SibFU
https://elib.sfu-kras.ru/handle/2311/154958

Creative Commons License This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0).