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