AI-Based Adaptive Personalized Learning Systems as a Mechanism for Enhancing Student Learning Outcomes in Digital Environments

Abstract

Traditional learning models prove insufficiently effective in ensuring the personalization of the educational process, which underscores the relevance of adaptive solutions capable of responding promptly to educational needs and learning behaviors of students in the digital environment. The purpose of this article is to substantiate the effectiveness of AI-based adaptive personalized learning systems (hereinafter APLS) as a mechanism for enhancing the quality of student learning outcomes in digital educational environments. Methods. The methodological foundation of the study comprises theoretical methods of analysis, synthesis, abstraction, induction, and deduction, which facilitated the systematization of scholarly approaches to the application of intelligent technologies in education. The empirical component was implemented through observation and description methods to examine the practical experience of deploying adaptive platforms in higher education institutions (hereinafter HEIs).

Results. The study established that the effectiveness of APLS is determined by the integration of machine learning technologies, neural networks, natural language processing, recommendation algorithms, and big data analytics tools that enable comprehensive processing of learning data. This allows for predicting the optimal difficulty level of learning materials, identifying common errors and patterns in student learning behavior, and dynamically adapting the content, pace, and formats of learning tasks. The use of the Moodle platform, Microsoft Teams, and foreign adaptive educational solutions (Knewton, DreamBox Learning, Carnegie Learning, ALEKS) was analyzed, which made it possible to characterize the level of readiness of the Ukrainian higher education system for the implementation of innovative digital technologies. It was found that the deployment of these platforms ensures an improvement in student learning outcomes within the range of 12 to 25 percent through the formation of individualized learning pathways, analytical support of the educational process, and provision of continuous feedback. Based on the obtained results, practical recommendations were developed for the implementation and optimization of adaptive systems, encompassing the development of digital infrastructure, integration with LMS, faculty training, monitoring of learning achievements, and ensuring personal data protection.

Conclusions. A pattern was identified whereby the enhancement of learning outcomes in the digital environment is directly dependent on the level of analytical processing of learning data and the degree of adaptation of educational content to the individual dynamics of students.