ESTIMATING STUDENT KNOWLEDGE LEVELS IN AN ITS FOR PROPOSITIONAL LOGIC

Abstract

This thesis tests the use of a student model in an intelligent tutoring system (ITS) for propo-sitional logic. ITSs are widely used in modern education programs and are applied for dif-ferent purposes. Student models are used in ITSs to determine if a student masters the subject of the ITS or how a student solves an exercise. Often, the student models are used to generate exercises that are relevant for the student.
In this thesis, we model and implement a student model that can track the knowl-edge levels of students on propositional rewrite rules. We will base the student model on Bayesian Knowledge Tracking, but extend the basic structure of the classical Bayesian Knowledge Tracking model. The student model will classify the difficulty level of the ex-ercises in the ITS based on the knowledge level of the individual students. We will gather feedback from the students on exercises in the ITS, build student models for each of the students, and compare the feedback with the classification of the student models.
We will show that the model works as intended, but that we need to refine the model to provide better classifications

Date of Award	20 Jun 2023
Original language	English
Supervisor	Bastiaan Heeren (Examiner) & Josje Lodder (Co-assessor)