AI-Driven Early Identification of At-Risk Students Through Learning Management System Behavioural Analytics

<p>The increasing availability of digital learning data has created new opportunities for educational institutions to identify students who may be at risk of poor academic performance before difficulties become difficult to reverse. This study investigates the use of behavioural analytics from Learning Management Systems (LMS) to support the early identification of at-risk students. Rather than relying solely on examination results or end-of-semester assessments, the proposed approach examines patterns embedded in students’ online learning activities, including login frequency, time spent on learning materials, assignment submission behaviour, assessment attempts, participation in discussion activities, and periods of inactivity. These behavioural indicators can be analysed to detect changes that may signal declining engagement or emerging academic difficulties. The study proposes an AI-driven framework that combines LMS data preprocessing, behavioural feature extraction, predictive modelling, and risk classification to generate timely information for academic intervention. Machine learning techniques can be employed to distinguish students who are likely to remain academically engaged from those whose behavioural patterns indicate an increased probability of underperformance or withdrawal. The framework is intended to assist lecturers, academic advisers, and student-support teams in moving from reactive intervention toward proactive and evidence-based support. At the same time, the use of student behavioural data requires careful consideration of privacy, fairness, transparency, data quality, and the potential consequences of inaccurate predictions. The study therefore emphasises responsible use of predictive analytics alongside appropriate human judgement. Overall, AI-supported LMS behavioural analytics offers a practical pathway for detecting early warning signals and enabling targeted interventions that can improve student engagement, retention, and academic outcomes.</p>