A Personalized Education Recommendation Algorithm Based on Student Learning Behavior and Graph Neural Networks

With the rapid development of online education, accurately capturing students’ learning behaviors and providing personalized learning recommendations has become a key issue in the field of educational technology. To address this challenge, this study proposes a Behavior-Guided Graph Learning Network (BGGLN) model, which integrates Long Short-Term Memory (LSTM) networks and Graph Neural Networks (GNN). The model utilizes LSTM to capture the temporal features of students’ learning behaviors, while GNN models the complex relationships between students and learning resources, enabling more precise personalized recommendations. This study evaluates the model using a multimodal dataset that includes student behavior data and a learning resource graph. Experimental results show that the BGGLN model outperforms traditional LSTM and GNN models in various metrics such as accuracy, F1 score, and R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> , achieving an accuracy of 93%, and effectively reducing errors in the recommendation system. Compared to other models, the BGGLN model demonstrates stronger learning capability and generalization performance, proving its practical application value in complex learning scenarios.