知识图谱与人工智能赋能的Python程序设计课程教学改革研究
摘 要：Python程序设计课程的教学长期面临三个问题：知识点零散，个性化指导不足，教师重复劳动负担重。针对这些问题，课程组将知识图谱与人工智能技术引入教学，经过一个完整学年的实践，形成了一套Python课程教学改革方案。先梳理出5个知识域、167个知识点节点和165条语义关系边，建成课程知识图谱；再开发AI辅助教学平台，把智能编程答疑、代码自动评阅、自适应习题推荐、个性化学习路径规划四项功能集成在一起。教学中按课前图谱定位、课中精讲突破、课后AI推荐拓深的顺序组织，形成混合式教学闭环。实践结果显示，学生的学习效果和自主学习能力都有明显提高，不及格率下降，学生对教学认可度高。这套做法对编程类课程教学改革有一定的参考价值，也具备推广条件。
关键词：知识图谱；人工智能；Python程序设计；教学改革；自适应学习；个性化推荐
正文
可下载并阅读全文PDF，请按照本文版权许可使用。
Download the full text PDF for viewing and using it according to the license of this paper.
Research on Teaching Reform of Python Programming Course Empowered by Knowledge Graph and Artificial Intelligence
Abstract: The teaching of Python programming courses for a long time has been plagued by three problems: knowledge points are scattered and personalized instruction is scarce, and instructors are burdened with lots of redundant tasks. In order to address these issues, our course team has introduced knowledge graph and artificial intelligence technologies in education. Based on one-year practical experience, we developed an improvement proposal of the Python course.We first organized the course’s 5 knowledge areas, 167 knowledge points as nodes, and 165 semantic relationship lines into a course knowledge map; then designed an artificial intelligence-aided teaching system integrating four capabilities of intelligent program answering,automatic code review, adaptive exercise recommendation, and personal learning plan. Teaching follows the following order: before class, the knowledge graph tells the student where he or she is; in the classroom, instructors provide targeted teaching key and hard points; after class, AI suggestions assist student to learn more deeply. It constitutes a closed-loop blended teaching process. Results show that students have improved in both learning performance as well as self-learning ability, the failure rate decreased and the student gave high praise to the teaching method. The method is worth referencing for teaching reform in programming courses, and it's ready for wider use.
Keywords: Large Language Model; AI Agents; Software Construction; Multi-Agent Collaboration; Curriculum Developmen
参考文献
- 蒋帆，刘思成，数字技术如何重构教师专业活动[J],教育发展研究 2026.04
- Wang Q, et al. Knowledge Graph Embedding: A Survey of Approaches and Applications[J]. IEEE TKDE, 2017, 29(12): 2724-2743.
- Brown T B, et al. Language Models Are Few-Shot Learners[C]. NeurIPS, 2020.
- Liu Q, et al. EKT: Exercise-Aware Knowledge Tracing for Student Performance Prediction[J]. IEEE TKDE, 2019.
- 李艳燕, 张香玲, 李新, 杜静. 面向智慧教育的学科知识图谱构建与创新应用[J]. 电化教育研究, 2019(8)...
- Piech C, et al. Deep Knowledge Tracing[C]. NeurIPS, 2015: 505-513.
- Zhu Y C, Zhang W, Chen Y, et al. RCD: Relation Map Driven Cognitive Diagnosis for Intelligent Education Systems[C]//SIGIR 2021: 1617-1627.
- 张昭玉, 胡伏原, 唐佳佳, 等. 基于课程知识图谱的个性化教学设计及实践[J]. 软件导刊, 2025, 24(11): 147-152
- Kazemitabaar M, Chow J, Ma C K T, Ericson B J, Weintrop D, Grossman T. Studying the Effect of AI Code Generators on Supporting Novice Learners in Introductory Programming[C]//ACM CHI 2023: 455:1-455:23.
- Wang Z, Alsam A, Morrison D, Strand K A. Toward Automatic Feedback of Coding Style for Programming Courses[C]//IEEE ICALT 2021: 56-60.
- Yeung C K, Yeung D Y. Addressing Two Problems in Deep Knowledge Tracing via Prediction-Consistent Regularization[C]//ACM L@S 2018: 158-167.
