Abstract:

In this work, we started from an existing Intelligent Tutoring System (ITS) called Sistema de Apoyo Generalizado para la Enseñanza individualizada (SAGE), able to supervise student’s learning according to the first four levels of Bloom’s taxonomy: knowledge, comprehension, application and analysis, and we propose to improve its reach by adding other functions according to Marzano’s taxonomy, that preserves the basic aspects of Bloom’s taxonomy and adds metacognition and emotional response. SAGE starts with a diagnostic test about the subject of study, to which a cognitive diagnostic test will be added, then the system assigns the lesson that must be completed, according to the previous knowledge of the student. While students navigate throughout the lesson, the system will monitor their emotional response and motivation using a camera, facial recognition and machine learning techniques. To decide which will be the next lesson a personalized advance route will be traced according to a student model, and to make the advance between lessons a shared control between the student and the system will be implemented. Using this methodology, teachers will be able to focus on activity planning and evaluation of assignments related with knowledge utilization, like essays or application projects, and the system will be in charge of the tasks of the remaining levels of Marzano’s taxonomy (retrieval, comprehension, analysis, metacognition and self-system).

Keywords:

Intelligent Tutoring Systems, Student Modeling, Affective Computing, Metacognition.