A Student Model for Teaching Natural Deduction Based on a Prover That Mimics Student Reasoning

João Carlos Gluz, Fabiane Penteado, Marcel Mossmann, Lucas Gomes, Rosa Vicari

Logic is a fundamental discipline for Computer Science, and Engineering students. However, despite its importance, there are several problems with the teaching of this discipline in graduate courses. Trying to improve this situation, we designed, and developed a new tutoring system for Logic, called Heraclito. This system implements a dynamic and adaptive student model, which is able to automatically solve the problems presented to students in a way similar to the employed by teachers, and, at the same time, is able to follow, and adapt itself to the form of reasoning used by students. The paper presents the main components of Heraclito’s student model, including the formal definition its similarity measurement function, and the similarity experiments conducted with Logic proofs generated by this system.

The final publication is available at Springer via https://doi.org/10.1007/978-3-319-07221-0_60.