Predicting Student Performance in Higher Education Based on Dynamic Graph Neural Networks with Consideration of Grading Habits

Accurately predicting student performance in higher education is crucial for educators and institutions to evaluate and improve teaching and learning outcomes. Traditional methods for predicting student performance often use machine learning algorithms that rely on the student learning behaviours, such as students’ performances of the past years or students’ attendance data, whereas the teacher grading habits are usually not considered yet. In this paper, we propose a new approach for predicting student performance in higher education based on dynamic graphical neural networks that consider not only the students behaviour but also teachers’ grading habits. We evaluate the proposed approach on a real-world dataset and show that our approach outperforms existing methods in terms of accuracy and F1-score.