Mathematical Apparatus for Some Quality Indicators Assessment of a Training Webinar Using Machine Learning

This article is the second stage of creating and researching a technology for assessing the quality of contact work at a university implemented through webinars. The mathematical support of the intelligent technology for analyzing the audio-video stream of an educational webinar recording is considered. The problem of increasing the efficiency and reliability of assessing some indicators of the quality of contact work at the Faculty of Distance Learning of SibSUTIS implemented through distance learning webinars is being solved. Machine learning methods and their implementation using standard Python modules and libraries are used for analysis. The scientific significance of the work lies in the development of new models and algorithms for intelligent analysis of audio-video streams parameterized for processing the recording of an educational event. The developed models and algorithms make it possible to increase the efficiency of the process of assessing the quality of contact work implemented through webinars.

1. Poletaikin A. N., Shevtsova Yu. V., Podkolzin V. V., Strukova E. G. Matematicheskaya model' otsenivaniya kachestva kontaktnoi raboty, realizuemoi posredstvom vebinarov v khode distantsionnogo obucheniya [Mathematical model for assessing the quality of contact work implemented through webinars during distance learning]. Informatika i obrazovanie, 2019, no. 7 (306), pp. 42-53, DOI: 10.32517/0234-0453-2019-34-7-42-53.

2. Sizov L. A. Vebinary – peredovaya tekhnologiya nepreryvnogo setevogo obrazovaniya [Webinars - advanced technology for continuous online education]. Problemy ekonomiki i informatizatsii obrazovaniya. Materialy XIV Mezhdunarodnoi nauchno-prakticheskoi konferentsii, Tula, TIEI, 2017. pp. 92-95.

3. Tsallagov N. A., Astakhova T. A. Provedenie effektivnykh vebinarov ot A do Ya [Conducting effective webinars from A to Z]. OOO «Mirapolis», 2015, 174 p., available at: http://indo.kg/wp-content/uploads/2018/11/Provedenie-effektivnykh-vebinarov-ot-A-do-Ya.pdf (accessed: 31.12.2023).

4. Sorokin G.A. Fil'try nizhnikh chastot [Low pass filters]. Vestnik YuUrGU, Seriya: Komp'yuternye tekhnologii, upravlenie, radioelektronika, 2015, no. 1, pp. 100-107.

5. Popov A. N., Shelupanov A. A. Mashinnoe obuchenie. Evolyutsiya metodov. [Machine learning. Evolution of methods]. Moscow, DMK Press, 2019. 304 p.

6. Gary Bradski, Adrian Kaehler. Learning OpenCV: Computer Vision with the OpenCV Library, First Edition, USA, O'Reilly Media, 2008. 543 p.

7. Andriyanov N. A., Dement'ev V. E., Tashlinskii A. G. Obnaruzhenie ob"ektov na izobrazhenii: ot kriteriev Baiesa i Neimana–Pirsona k detektoram na baze neironnykh setei EfficientDet [Detection of objects in an image: from Bayes and Neyman-Pearson criteria to detectors based on EfficientDet neural networks]. Komp'yuternaya optika, 2022, vol. 46, no. 1, pp. 139-159. DOI: 10.18287/2412-6179-CO-922.

8. Kazemi V., Sullivan J. One millisecond face alignment with an ensemble of regression trees. 2014 IEEE Conference on Computer Vision and Pattern Recognition, Columbus, OH, USA, 2014, pp. 1867-1874, DOI: 10.1109/CVPR.2014.241.

9. Liang J., Doermann D., Li H. Camera-based analysis of text and documents: a survey. IJDAR, 2005, vol. 7, pp. 84-104, DOI: 10.1007/s10032-004-0138-z.

10. Dvoryankin D. K., Kuleshov D. A., Kadurin A. V. Neironnye seti i glubokoe obuchenie [Neural networks and deep learning]. Saint Petersburg, Piter, 2019. 352 p.