Student Performance

Predicting student performance is a key area of educational research, aiming to identify at-risk students and improve learning outcomes. Current research focuses on developing predictive models using diverse data sources, including learning management system logs, physiological signals, and even AI-generated think-aloud transcripts, employing machine learning algorithms like support vector machines, random forests, and transformer networks. These models, while showing promising accuracy in predicting grades and identifying at-risk students, are also being rigorously evaluated for algorithmic bias and refined to incorporate temporal dynamics and multi-modal data for improved accuracy and generalizability. The ultimate goal is to leverage these insights to create more effective and personalized learning interventions.