Stručni rad
https://doi.org/10.15516/cje.v28i3.42892
Application of Machine Learning Technologies for Personalising the Digital Educational Process in Higher Education Institutions
Nurkylych Nuraliev
; Department of Pedagogy Kyrgyz State University named after I. Arabaev
*
Piotr Kusznieruk
; Faculty of Social Sciences Warsaw Medical Academy
Aizada Azhibekova
; Department of Automated Systems and Digital Technologies Osh State University
Andrii Burachynskyi
; Educational and Scientific Institute of Information Technology State University of Information and Communication Technologies
Hristo Kamenov
; Paisii Hilendarski University of Plovdiv
* Dopisni autor.
Sažetak
Abstract
The aim of the study was to identify the main types of machine learning (ML) models that help adapt the educational environment to the individual needs of students in universities. The methodology combined a systematic analysis of approaches to teaching using ML algorithms with a comparative analysis of the digital transformation of higher education in Kyrgyzstan, Azerbaijan, Poland, and Ukraine. The analysis of ML approaches and the comparative analysis of the digitalisation of the examined educational institutions revealed key mechanisms for personalising the learning process. The comparison of practices showed that Poland and Ukraine prefer reinforcement learning models and academic performance prediction algorithms; Azerbaijan used clustering methods to group students by learning styles; and Kyrgyzstan implemented recommendation systems to improve motivation and feedback. The analysis suggests that the implementation of these technologies depends on four key factors: the level of digital competence of teachers, the availability and quality of educational data, the integration of adaptive modules into learning platforms, and the presence of state policies regulating the use of artificial intelligence (AI) in education. Additionally, it was found that personalised ML modules contribute to increasing students' intrinsic motivation, the development of self-learning skills, and active interaction within learning groups. The practical value of the work lies in the recommendations for deploying adaptive ML modules in higher education institutions in the countries under study, which will help improve the quality of education, student motivation, and reduce the gap in academic achievements.
Keywords: analytics; artificial intelligence; educational behaviour analysis; platforms; motivation; social interaction
Ključne riječi
analytics; artificial intelligence; educational behaviour analysis; platforms; motivation; social interaction
Hrčak ID:
351227
URI
Datum izdavanja:
21.9.2026.
Posjeta: 0 *