RF03-07 SAAT - 1: Academic Analytics and Early Warning System Details

Start Date: 2026-06-08

End Date: 2026-06-08

Higher education institutions face the ongoing challenge of identifying students at risk of academic delay, poor performance, or dropout at an early stage. Traditional approaches are typically based on isolated indicators and retrospective analyses, which often hinder timely intervention. This paper proposes the Academic Analytics and Early Warning System (SAAT), a dynamic vectorial mathematical model that transforms academic observations into analytical indicators and subsequently into risk metrics, enabling the construction of temporal trajectories for each student and the generation of explainable alerts. The model defines three interrelated vector spaces: • Observed Variables Space. • Analytical Indices Space. • Academic Risk Space. The proposed architecture allows the complete reconstruction of a student's academic state at any point in time and measures its evolution through change functions specifically designed for educational analytics.