Personalized Learning Path Generation Algorithm Based on Graph Neural Networks

With the rapid development of information technology and educational resources, traditional learning path recommendation systems are often based on students’ historical data or resource categorization, making it difficult to fully meet the personalized needs and ability differences of students. This study proposes a personalized learning path generation algorithm based on Graph Neural Networks (GNNs), which can recommend the most suitable learning paths for students based on their abilities, interests, and learning needs, thereby improving learning efficiency and optimizing the allocation of educational resources. This paper first constructs the model by introducing the concept and application of GNNs, then conducts experimental comparisons with content recommendation algorithms and collaborative filtering algorithms, and finally validates the effectiveness of the proposed algorithm on a specific dataset. The experiments are conducted on a dataset of 50 students from an internet education platform. The model training utilizes students’ behavioral data, interests, hobbies, and study time for training and fine-tuning. The experimental results show that the algorithm proposed in this study outperforms traditional content recommendation and collaborative filtering algorithms in terms of recommendation accuracy, recall rate, and F1 score. This research provides theoretical and practical guidance for the design of personalized learning paths in the direction of online education.