Understanding Student Success Prediction Using SHapley Additive exPlanations

Predicting student success is an important task in educational institutions, as it allows for targeted interventions and support systems to enhance educational outcomes. This paper explores the use of SHAP (SHapley Additive exPlanations) model-agnostic method in understanding and interpreting student success prediction. The predictive model was built using Multi-Layer Perceptron neural network algorithm on a large public dataset. By shedding light on the underlying factors driving student success, this research contributes to the advancement of data-driven decision-making in education.