“An interpretable hybrid ensemble model for early academic risk detection with shap-based explainability” (2026) Journal of Soft Computing Exploration, 7(3), pp. 489–501. doi:10.52465/joscex.v7i3.50.