Design and Practice of an Intelligent Nursing Auxiliary Teaching System Based on Knowledge Graph

Chen Zhoujing

Hainan Vocational University of Science and Technology

Abstract：

Nursing education is currently undergoing a structural transformation toward digital and intelligent paradigms.However, existing instructional platforms frequently struggle to map the complex, non-linear relationships among clinical symptoms, pathological mechanisms, and nursing interventions.This structural deficit impedes students from developing holistic clinical reasoning skills.To mitigate this problem, this paper proposes the design and implementation of an intelligent nursing auxiliary teaching system driven by a domain-specific Knowledge Graph (KG). First, we define a hierarchical nursing ontology based on standard medical nursing textbooks and clinical guidelines.Subsequently, a deep learning framework comprising RoBERTa-WWM-ext, Bidirectional Long Short-Term Memory (BiLSTM), and Conditional Random Fields (CRF) is deployed to extract medical entities, while Piecewise Convolutional Neural Networks (PCNN) are utilized for relation extraction.The extracted triples are stored in a Neo4j graph database, establishing a comprehensive nursing knowledge network. Building upon this infrastructure, the system provides three core pedagogical functionalities: semantic-based intelligent question-answering, personalized learning path recommendation, and graph-driven clinical case simulation.An empirical study involving 80 undergraduate nursing students demonstrates that the experimental group utilizing the KG-based system achieved significantly higher scores in both theoretical examinations and clinical thinking assessments compared to the control group (p < 0.05). The findings confirm that integrating knowledge graphs into nursing curricula effectively bridges the gap between theoretical knowledge and clinical practice, offering a scalable solution for smart medical education.

Key Words：

Knowledge Graph; nursing education; intelligent teaching system; natural language processing; clinical reasoning; smart education