Application of ChatGPT-like Large Language Model in nursing course assessment—Based on ChatGPT,ERNIE Bot,iFLYTEK SPARK test
Objective To explore the application effect of ChatGPT-like Large Language Model in nursing course assessment.Methods Three Large Language Models(ChatGPT,ERNIE Bot and iFEKLYT SPARK)were tested by using the test of"Basic Knowledge of Nursing"and the exam of"Basic Nursing(undergraduate)",and the scores were compared with students'scores.Results In terms of acquisition of basic nursing knowledge,the ability of the three Large Language Models was similar with more than 90%of the accuracy rate.However,in the application ability of nursing knowledge,the accuracy rate of the three Large Language Models was relatively low and dif-ferences existed.In addition,the Large Language Models performed well in terms of speed of content generation,ease of use,and limitation of use.Conclusion ChatGPT-like Large Language Model has advantages in solving basic nursing knowledge problems,but has limitations in problems requiring knowledge application or independent thinking.With the constant development of technology,the new generation of Large Language Models is expected to promote the innovation of test ordering,in-telligent evaluation,and teaching effect evaluation.
万方数据
