Research on Classification of Medical VR Resources Based on Knowledge Graph

This paper proposes an automated classification method based on knowledge graph, aiming to address the classification and management challenges brought by the rapid growth of medical virtual reality (VR) resources. A classification algorithm was designed by constructing a medical knowledge graph and ex-tracting multimodal features of VR resources, which integrates text, image, and user interaction data. Experimental results show that the method has signifi-cantly improved accuracy, recall, and F1 value compared to traditional methods, proving the effectiveness of knowledge graph in enhancing classi-fication performance. In addition, the author has developed a personalized recommendation system that provides customized resource recommendations based on users' learning needs, significantly improving the learning experience and effectiveness. Providing innovative tools for future medical education and clinical practice teaching, it also provides valuable reference for the devel-opment of medical VR resource management technology.