Abstract:

Objective To construct a knowledge graph for the Internal Medicine Nursing course and explore its preliminary application effects among undergraduate nursing students. Methods The knowledge graph was developed through ontology construction，knowledge extraction，knowledge fusion，knowledge storage and visualization. From September 2024 to January 2025，it was initially applied to the Internal Medicine Nursing course for 83 third-year undergraduate nursing students. Post-implementation，students’ autonomous participation rates，knowledge point completion rates，and mastery rates were evaluated. Using an 80% knowledge point completion rate as the threshold，students were divided into high-completion and low-completion groups. Course performance was compared between these groups，and focus group interviews were conducted to gather students’ perceptions and suggestions regarding the knowledge graphs. Results The constructed knowledge graph encompassed 1 668 knowledge points，70 integrated teaching videos，and 364 self-test questions. The autonomous participation rate，and mean values for students’ knowledge point completion rate and mastery rate were 60.24%，45.27%，and 30.34%，respectively. After matching 21 students from the low-completion group（n=62） with the high-completion group（n=21） based on prior course performance and basic characteristics，the high-completion group scored significantly higher than the 21 students from the low-completion group in Internal Medicine Nursing[（77.35±8.94） vs. （70.08±12.22） points，t=2.201，P=0.034]. Interviews with 23 students（11 high-completion and 12 low-completion） revealed three themes：benefits gained from participation，factors hindering participation，and urgent improvements needed for the knowledge graph and its platform. Conclusion The knowledge graph of Internal Medicine Nursing course can improve students’ learning effect to a certain extent，but it still needs to be further optimized and improved based on artificial intelligence and other technologies to develop a precise learning model grounded in course knowledge graph.

Key words: Education, Nursing, Internal medicine nursing, Knowledge graph, Artificial intelligence, Precision learning