AI Supported Learning Analytics for Student Performance Prediction

Abstract

Artificial Intelligence (AI)-supported learning analytics has emerged as a transformative approach for predicting student performance and improving educational decision-making in higher education. By combining learning analytics with advanced Artificial Intelligence techniques such as Large Language Models (LLMs), machine learning, and predictive modeling, educational institutions can identify learning patterns, forecast academic outcomes, and provide timely interventions for students at risk of poor performance. Traditional learning analytics systems often rely on descriptive statistics and historical academic records, limiting their ability to deliver personalized and real-time educational support. AI-supported learning analytics overcomes these limitations by integrating curriculum-aware artificial intelligence, Retrieval-Augmented Generation (RAG), Explainable Artificial Intelligence (XAI), Human-in-the-Loop (HITL) verification, and institutional governance frameworks to improve prediction accuracy and educational transparency. This study investigates AI-supported learning analytics for student performance prediction through a qualitative review of recent scholarly literature. The findings indicate that AI-enhanced learning analytics significantly improve prediction accuracy, learner engagement, personalized learning, curriculum alignment, and institutional decision-making while reducing dropout risks and improving academic success. Although implementation requires technological infrastructure, continuous curriculum updates, data governance, and faculty development, AI-supported learning analytics provides a sustainable framework for intelligent, ethical, and learner-centered educational environments.