Artificial Intelligence and Learning Analytics for Comparative Physics Education: Bridging the Didactic Gap between Moroccan and Scandinavian Higher Education

Abstract

Physics is widely recognized as one of the most cognitively demanding disciplines, requiring students to integrate abstract mathematical reasoning with the interpretation of invisible physical phenomena. While Nordic higher education systems have successfully addressed this complexity through student-centered learning, formative assessment, and phenomenon-based pedagogy, massified higher education systems such as Morocco continue to face significant challenges related to large class sizes, limited individualized support, and linguistic barriers.

This study proposes a comparative didactic framework that integrates Artificial Intelligence (AI), Educational Data Mining (EDM), and Learning Analytics (LA) to support physics teaching in resource-constrained university environments. The proposed framework focuses on complex electromagnetism concepts and combines interactive virtual simulations with unsupervised machine learning techniques to analyze students' learning behaviors. A K-Means clustering model is employed to identify distinct misconception profiles from learners' interaction logs, enabling instructors to detect cognitive difficulties at an early stage and deliver targeted pedagogical interventions.

Rather than replacing teachers, the proposed AI-supported architecture extends formative assessment capabilities by providing continuous diagnostic feedback that would otherwise be impractical in large university classes. The framework also illustrates how learning analytics can reproduce key pedagogical characteristics commonly associated with Nordic educational practices while remaining compatible with the structural constraints of Moroccan higher education.

This work contributes to comparative physics education by establishing a conceptual bridge between educational systems through AI-supported learning analytics and offers a scalable framework for improving conceptual understanding, adaptive instruction, and evidence-based teaching in complex STEM disciplines.

Keywords: Physics Education; Comparative Didactics; Artificial Intelligence in Education; Learning Analytics; Educational Data Mining; Electromagnetism; Higher Education; Morocco; Nordic Education.

Received Date: 20 October 2025 

Accepted Date: 11 November 2025 

Published Date: 1 December 2025

Available Online at: https://www.ijsrisjournal.com/index.php/ojsfiles/article/view/766