This paper presents EduPredict-X+, a novel machine learning framework for predicting academic risk among students using the UCI Student Performance Dataset. The model incorporates counterfactual analysis and elasticity-driven insights to understand how changes in student behavior and performance indicators impact outcomes. Unlike traditional predictive models, EduPredict-X+ not only identifies at-risk students but also provides interpretable insights into the contributing factors. The framework is designed to assist educators and institutions in making data-driven decisions for early intervention. The proposed approach demonstrates improved predictive accuracy and interpretability compared to baseline models. This work contributes to the growing field of AI in education by combining predictive analytics with explainable AI techniques. Keywords: Academic Risk Prediction, Machine Learning, Explainable AI, Counterfactual Analysis, Student Performance.
Prajakta Singhal (Fri,) studied this question.