EduRiskX combines neural and symbolic AI to predict academic risk earlier and more transparently

Researchers propose EduRiskX, a neuro-symbolic framework that integrates a temporal Transformer-based predictor with F-Logic symbolic reasoning to predict students' academic risk in online education.

The neural component models longitudinal student activity sequences using temporal attention, class-weighted loss, and dynamic weekly truncation, while the F-Logic rule base is grounded in established educational theories (Engagement Theory and Student Integration Model) and constructed exclusively from training data.

Experiments on the Open University Learning Analytics Dataset (OULAD) using a strict 80/10/10 student-level split show that EduRiskX achieves an accuracy of 0.900 and an F1-score of 0.894 by the end of the semester (week 38), with an average early detection week of 9.32 and a detection rate of 94.30%.

Under identical conditions, EduRiskX yields improved recall and earlier risk identification compared with state-of-the-art time-series models (PatchTST, iTransformer) and common deep learning baselines (LSTM, CNN).

Beyond predictive performance, the F-Logic module provides structured rule-based explanations that link predictions to observable behavioral patterns and educational theories, addressing the interpretability challenges that hinder adoption of black-box models in educational settings.