Imagine a classroom where every student receives a customized learning experience, perfectly tailored to their individual needs and learning style. This is the promise of personalized learning (PL), and artificial intelligence (AI) is poised to make it a reality, especially in STEM subjects. While AI has been transforming education at all levels, a recent study from Universiti Teknologi Malaysia (UTM) reveals a gap in research focusing on AI-driven personalized learning specifically for pre-university students.
This comprehensive review, spanning the last decade (2013-2023), analyzed 69 research articles from the Scopus database to understand how AI is being used to personalize learning. The study looked at AI techniques, the elements of personalized learning considered, and the potential for replicating these approaches at the pre-university level. The research team at UTM sought to identify gaps and opportunities for future studies in this vital area. For instance, AI has already made significant strides in areas such as adaptive testing and personalized feedback systems. However, these advances have not been fully translated into personalized learning experiences at the pre-university level.
The UTM study found that while machine learning and deep learning are the most commonly used AI techniques, most existing research focuses on higher education. The elements most needed for personalized systems include knowledge delivery methods and understanding each learner’s needs, behavior, and interests. The study also highlighted that Switzerland, the USA, the UK, and China are currently leading the charge in PL research output.
The significance of this research lies in its potential to revolutionize STEM education. By identifying the gaps in current research, the UTM study paves the way for developing AI-driven personalized learning systems that can significantly enhance learning outcomes and satisfaction for pre-university students. This could translate to a more engaged and better-prepared generation of STEM professionals. The next step is to thoughtfully integrate educational theories, subject-specific content, and industry needs into these AI-driven systems to ensure they are both effective and relevant. The findings of this systematic literature review offers insights that can guide future investigations and development efforts in creating AI-driven personalised learning systems that enhance the outcomes for pre-university STEM students.
DOI: https://doi.org/10.14569/ijacsa.2025.0160636
