Predicting Online Learning Performance and Designing Interventions Using Learning Analytics

With the widespread adoption of online learning, educators face the challenge of effectively improving student performance. This study aims to establish a grade prediction model and design corresponding learning intervention strategies by analyzing online learning behaviors. We utilized learning data from the edX online learning platform, including study duration, engagement, and quiz scores. Spiking neural network was employed to analyze the data and identify key behavioral features that influence student performance. The results indicate a significant correlation between learning behaviors and grades, with specific study habits and engagement levels playing a crucial role in grade prediction. Furthermore, based on the analysis results, we designed and implemented a series of learning intervention measures. This research provides empirical evidence for personalized learning and teaching interventions, emphasizing the importance of data-driven decision-making in education.