Détection et Analyse de l’Engagement des Étudiants à l’aide du Suivi Oculaire

Abstract

Understanding learner engagement is a key principle in education. It helps adapt teaching methods and educational content to maximize the effectiveness and success of each student. Various techniques have been proposed in the literature. In this work, we focus on integrating the eye tracking model as a tool for extracting different ocular metrics to detect learner engagement. The first objective of this work is to propose a structure for presenting educational content. This structure aims to maximize the amount of relevant information captured by the model. The second objective is to detect moments of learner engagement during the learning process. Machine learning algorithms, such as Random Forest and XGBoost, have been applied to detect engagement in real-time. All these proposals have been implemented in our E-track Learning system. This study opens new perspectives for personalized teaching and continuous content adaptation, enabling real-time feedback and more tailored learning experiences for students.