This paper proposes a novel model, APP-DGNN, for predicting academic performance using dual graph neural networks. The model combines interaction-based and attribute-based graph neural networks to capture both the structural information from online learning activities and the attribute features of students. The interaction-based graph neural network (IGCN) learns local academic performance representations from online interaction activities, while the attribute-based graph neural network (AGCN) learns global academic performance representations from student attributes using dynamic graph convolution operations. The learned representations from local and global levels are combined in a local-to-global representation learning module to generate predicted academic performances. The empirical study results demonstrate that the proposed model significantly outperforms existing methods. Notably, the proposed model achieves an accuracy of 83.96% for predicting students who pass or fail and an accuracy of 90.18% for predicting students who pass or withdraw on a widely recognized public dataset. The ablation studies confirm the effectiveness and superiority of the proposed techniques. The model is evaluated on a public dataset from the Open University Learning Analytics dataset (OULA), and the results show that the proposed model outperforms several baseline models, including traditional machine learning models and graph-based models. The model's performance is evaluated using classification accuracy and F1-score metrics. The results show that the proposed model achieves high accuracy in predicting academic performance, demonstrating its effectiveness in capturing complex relationships in online learning activities and student attributes. The model's architecture and training process are detailed, and the results show that the model is able to capture students' academic states from learning behavior data. The model's performance is also evaluated on different training sizes, showing that it maintains better performance with different sizes of training sets. The results confirm the effectiveness of the proposed model in predicting academic performance.This paper proposes a novel model, APP-DGNN, for predicting academic performance using dual graph neural networks. The model combines interaction-based and attribute-based graph neural networks to capture both the structural information from online learning activities and the attribute features of students. The interaction-based graph neural network (IGCN) learns local academic performance representations from online interaction activities, while the attribute-based graph neural network (AGCN) learns global academic performance representations from student attributes using dynamic graph convolution operations. The learned representations from local and global levels are combined in a local-to-global representation learning module to generate predicted academic performances. The empirical study results demonstrate that the proposed model significantly outperforms existing methods. Notably, the proposed model achieves an accuracy of 83.96% for predicting students who pass or fail and an accuracy of 90.18% for predicting students who pass or withdraw on a widely recognized public dataset. The ablation studies confirm the effectiveness and superiority of the proposed techniques. The model is evaluated on a public dataset from the Open University Learning Analytics dataset (OULA), and the results show that the proposed model outperforms several baseline models, including traditional machine learning models and graph-based models. The model's performance is evaluated using classification accuracy and F1-score metrics. The results show that the proposed model achieves high accuracy in predicting academic performance, demonstrating its effectiveness in capturing complex relationships in online learning activities and student attributes. The model's architecture and training process are detailed, and the results show that the model is able to capture students' academic states from learning behavior data. The model's performance is also evaluated on different training sizes, showing that it maintains better performance with different sizes of training sets. The results confirm the effectiveness of the proposed model in predicting academic performance.