| 基金项目:首都医科大学教育教学改革研究课题（2022JYY193） |
|
| 摘要点击次数: 276 |
| 全文下载次数: 90 |
| 中文摘要: |
| 目的 以“超声在麻醉、疼痛和重症医学中的应用”课程为例，构建临床医学专业课程学习者画像模型。方法 梳理临床医学专业课程学习者画像模型构建框架及流程，对多来源学习者数据进行收集及预处理。利用Python 3.9编程语言进行统计分析、自然语言处理、聚类分析，并以可视化技术呈现，构建机器学习预测模型并评估模型预测效能。结果 从学习背景、兴趣偏好和行为效果3个维度构建临床医学专业课程学习者画像模型。学习背景画像揭示了学习者基本信息、认知基础和学习动机。兴趣偏好画像分析了学习目的与其他选课信息，根据内容关注程度、学习效果影响因素识别出3类不同的学习者群体。行为效果画像通过构建4种机器学习算法预测模型实现了对课程考试成绩的分类预测，结果显示朴素贝叶斯算法效果最佳，准确率为0.80、F1分数为0.79，受试者工作特征（receiver operating characteristic，ROC）曲线下面积（area under the curve，AUC）为0.79（P=0.035），与其他算法差异有统计学意义。结论 本研究构建了临床医学专业课程学习者画像模型并开展实证研究，画像结果对教学内容、教学方式与教学团队、学习效果预测提供了指导建议。 |
| 英文摘要: |
| Objective To construct a learner profile model for clinical medical courses with the course "Application of ultrasound in anesthesia, pain and critical care medicine" as an example.Methods The framework and procedures were established for constructing the learner profile model for clinical medicine courses, and learner data from multiple sources were collected and pretreated. Python 3.9 programming language was used to perform statistical analysis, natural language processing, and cluster analysis, and the results were presented using visualization techniques. Machine learning prediction models were established and evaluated in terms of prediction performance.Results The learner profile model for clinical medical courses was established from the three dimensions of learning background, learning interests and preferences, and learning behaviors and effects. The learning background profile revealed the basic information, cognitive foundation, and learning motivation of learning. The interests and preferences profile analyzed the learning objectives and other selective courses of learners, and three different groups of learners were identified based on content attention and influencing factors for learning effects. The behaviors and effects profile achieved the classified prediction of course examination scores by constructing four machine learning algorithms, and the results showed that the naive Bayes algorithm had the best effect, with an accuracy of 0.81, an F1 score of 0.79, an area under the receiver operating characteristic curve of 0.79, and a P value of 0.035, indicating significant differences between the naive Bayes algorithm and other algorithms.Conclusion This study constructs a learner profile model for clinical medical courses and conducted an empirical study, and the results of this model provides guidance and suggestions for teaching contents, teaching methods, teaching team, and learning effect prediction. |
|
查看全文 查看/发表评论 下载PDF阅读器 |
|
| 关闭 |
|
|
|
