Machine Learning Algorithm for Real-Time Feedback on Mobile Learning Platforms for Students in High School

This study explores the application of machine learning algorithms to deliver real-time feedback in mobile learning platforms for high school students. Previous research has shown that real-time feedback can significantly improve student engagement and academic performance, particularly when supported by predictive models such as Neural Networks (NNs) and Decision Trees (DTs). This study addresses the following research question: Can machine learning models accurately predict student engagement levels to provide timely and personalized feedback in mobile learning environments. It is hypothesized that NNs will outperform DTs in predictive accuracy due to their ability to model complex, non-linear relationships. Using an experimental field study design, the research involved 150 high school students aged 16–18, who interacted with an AI-integrated mobile learning platform. The collected data included demographic attributes, platform usage, quiz performance, and engagement levels. Results showed that the NN model achieved 85% accuracy, outperforming the DT model (82%), with higher precision, recall, and F1-score values. These findings suggest that machine learning can effectively support personalized learning through adaptive feedback systems.