D.D. Matorin1, A.Yu. Cherepkov2, D.S. Zaitsev3

1–3 Bunin Yelets State University (Yelets, Russia)
 

The problem to which the article is devoted is the development of effective neural network architectures for an objective and consistent assessment of students' knowledge on a given scale based on quantitative data. An important area of research is the development of an approach to the creation and implementation of neural networks that allow efficient processing of pedagogical data and predicting objective assessments of knowledge acquisition using multilevel structures and logistic regression. The article studies approaches to the development of neural network architectures for modeling educational processes. The aim of the study is to develop effective models for the analysis of pedagogical data using neural networks, multilevel structures and multinomial logistic regression. The use of neural network architectures for modeling educational processes is considered. The case of an agent system for the formation of a module for assessing students' knowledge is studied. A neural network architecture is proposed for modeling a multilevel knowledge assessment system. The use of multinomial logistic regression in the framework of neural network modeling of the learning environment is considered. An example of the implementation of a multinomial logistic regression model is given, which can be used to predict a student's belonging to each of the four assessment classes based on input characteristics. The obtained results can be used in the problems of predicting student progress, in the problems of developing adaptive learning systems taking into account personalization, as well as in the problems of modeling educational processes.

Matorin D.D., Cherepkov A.Yu., Zaitsev D.S. Neural network architectures for modeling of educational processes. Neurocomputers: development, application. 2023. T. 25. № 1. С. 63–71. DOI: https://doi.org/10.18127/ j19998554-202301-03 (In Russian)

1. Panyukova S.V. Informacionnye i kommunikacionnye tekhnologii v lichnostno-orientirovannom obuchenii. M.: Progress. 1998. 205 s. (In Russian).
2. Klarin M.V. Pedagogicheskie tekhnologii v uchebnom processe. M.: Znanie. 2008 (In Russian).
3. Vaj Yan Min. Primenenie nejronnyh setej dlya kontrolya i prognozirovaniya rezul'tatov uchebnogo processa v vuze. Materialy mezhdunarodnoj nauchno-prakticheskoj konferencii «Chetvertye dekartovskie chteniya «Racionalizm i universalii kul'tury». Chast' 2. M.: Nacional'nyj issledovatel'skij universitet «Moskovskij institut elektronnoj tekhniki». 2017. S. 213–218 (In Russian).
4. Druzhinina O.V., Korepanov E.R., Belousov V.V., Masina O.N., Petrov A.A. Opyt razrabotki metodov i sredstv nejrosetevogo modelirovaniya nelinejnyh sistem na baze otechestvennoj vychislitel'noj platformy «El'brus 801-PC». Nelinejnyj mir. 2020. T. 18. № 2. S. 5–17. DOI https://doi.org/10.18127/j20700970-202002-01 (In Russian).
5. Chernomordov S.V., Druzhinina O.V., Masina O.N., Petrov A.A. Primenenie metodov mashinnogo obucheniya v zadachah nejrosetevogo modelirovaniya upravlyaemyh tekhnicheskih sistem. Nejrokomp'yutery: razrabotka, primenenie. 2022. T. 24. № 1. S. 25–35. DOI https://doi.org/10.18127/j19998554-202201-03 (In Russian).
6. Hajkin S. Nejronnye seti: polnyj kurs. M.: Vil'yams. 2006.
7. Ruanet V.V., Hetagurova A.K. Nejrosetevye tekhnologii kak sredstvo organizacii obrazovatel'nogo processa. Obrazovatel'nye tekhnologii i obshchestvo. 2005. № 4. S. 296–317 (In Russian).
8. Gricenko E.M., Osavelyuk P.A. Metod nejrosetevogo modelirovaniya postroeniya individual'nyh traektorij obucheniya. Vestnik PGU im. Sholom-Alejhema. 2013. №1(12). S. 58–67 (In Russian).
9. Grushevskij S.P., Dobrovol'skaya N.Yu., Kol'cov Yu.V. Organizaciya uchebnogo processa na osnove nejrosetevoj komp'yuternoj obuchayushchej sistemy. Vestnik Adygejskogo gosudarstvennogo universiteta. Seriya 3: Pedagogika i psihologiya. 2008. № 7. S. 142–148 (In Russian).
10. Petrov A.A., Druzhinina O.V., Masina O.N. Modelirovanie sistem ocenivaniya znanij v ramkah gibridnoj intellektual'noj obuchayushchej sredy. Sovremennye informacionnye tekhnologii i IT-obrazovanie. 2021. T. 17. № 1. S. 1–14 (In Russian).
11. Grigor'ev A.P., Mamaev V.Ya. O primenenii nejronnyh setej v testirovanii znanij. Nauchnoye priborostroenie. P. 2016. № 4. S. 77–84 (In Russian).
12. Hsiang-Fu Yu, Fang-Lan Huang, Chih-Jen Lin. Dual coordinate descent methods for logistic regression and maximum entropy models. Machine Learning. 2011. V. 85. P. 41–75.