Using Explainable AI to Rank Educational Success Factors Driving the Job Market

This research explores how Explainable AI (XAI) and Machine Learning (ML) enhance educational models to improve student performance and engagement, particularly in applying knowledge for job readiness. Analysing data from the graduates’ database, the study evaluates key factors such as viewership, participation, absenteeism, learning perception and study habits. ML models predict student outcomes in five categories (A, B, C, D and F), incorporating SHapley Additive exPlanations (SHAP) for transparency. Results show high-performing students (Class A) excel in soft skills (0.03980). Still, they are less influenced by family qualification (–0.00214), while low-performing students (Class F) score poorly in knowledge (0.00073) and are negatively affected by TikTok (–0.05253) and social media (–0.00452). Findings highlight the importance of improving synthesis skills, soft skills and family support while reducing digital distractions to bridge performance gaps. This study demonstrates how XAI and ML can guide targeted interventions to enhance learning and job preparedness.