AI-Driven Personalized Microlearning Framework for Enhanced E-Learning

Abstract

There has been increased demand for personalized approaches for e-learning that seek to increase the learners' engagement and outcomes over the past years. This has been triggered by the availability of mobile technologies and the exigence for adaptive instructional models that tailor the learning content to the learner's needs and settings. Microlearning, as an emerging paradigm of e-learning, is an original instructional approach that delivers time-efficient content that is provided to learners on demand. Microlearning can benefit a great deal from AI techniques to adapt the learning content to a variety of learners. This study proposes AI-driven personalized microlearning e-courses for higher education, especially for computer science courses. In this study, we develop and evaluate AI algorithms to produce adaptive learning paths for individual students, according to the data from the Open University Learning Analytics Dataset. Unlike existing approaches that rely on static, one size fits all instructional platforms, AI algorithms learn dynamically, predict and react to specific student needs to a fidelity of over 98% as shown in the experiments done in this study where their performance reached 98.96% accuracy, 99% precision and 99% F1-Score, and actually point to the use of highly tailored learning experiences to enhance both engagement and academic success. This contribution to the body of research on AI applications in education and on the potential for AI in improving personalized learning in computer courses is pointed out. Additionally, the study paves the way to embed adaptive microlearning strategies within current Virtual Learning Environments to address the individual learning requirements of students in today's digital classrooms.