大学化学

上一篇    下一篇

知识图谱与AI助教双轮驱动分析化学教学模式构建与实践

崔莹1, 马玉花2   

  1. 1 新疆理工职业大学通识学院, 新疆 喀什 844004;
    2 新疆师范大学化学化工学院, 新疆 乌鲁木齐 830054
  • 收稿日期:2026-07-13 录用日期:2026-08-27
  • 通讯作者: 马玉花 E-mail:15199141253@163.com Yuhua Ma
  • 基金资助:
    新疆生产建设兵团职业教育教学改革研究一般项目(BTBKXM-2025-Y123)；新疆维吾尔自治区高校本科教育教学研究和改革项目(XJGXPTJG-202366)

Construction and practice of a dual-drive analytical chemistry teaching model based on knowledge graph and AI teaching assistant

Ying Cui1, Yuhua Ma2   

  1. 1 School of General Education, Xinjiang Polytechnic University of Technology, Kashgar 844004, Xinjiang Uygur Autonomous Region, China;
    2 School of Chemistry and Chemical Engineering, Xinjiang Normal University, Urumqi 830054, Xinjiang Uygur Autonomous Region, China
  • Received:2026-07-13 Accepted:2026-08-27
  • Contact: Yuhua Ma E-mail:15199141253@163.com

PDF

10

赞

摘要： 针对分析化学课程知识点关联复杂、学生学习过程碎片化、学情反馈滞后、个性化支持不足及课程知识图谱“建而不用”等问题，构建知识图谱与AI助教双轮驱动的教学模式。该模式以课程知识图谱呈现知识结构、引导学习路径并诊断薄弱节点，以AI助教开展即时答疑、资源推荐和针对性训练，形成“诊断-干预-训练-反馈”的学习支持闭环。依托超星学习通平台，建设覆盖11个章节、567个核心知识点和91项教学资源的分析化学知识图谱，并基于1000余条课程问答语料训练AI助教。课程实践采用实验班与对照班描述性比较，并结合问卷反馈分析应用效果。结果显示，实验班课后作业平均成绩较对照班提高1.6% (P = 0.049 ＜ 0.05)，满分人数增加8名，满分率提高17.2% (P = 0.037 ＜ 0.05)；69名学生问卷中满意及以上比例为91.3%。相关结果为分析化学课程数智化教学中知识图谱与AI助教协同应用提供了实践案例。

关键词: 知识图谱, AI助教, 分析化学, 混合式教学, 数智赋能

Abstract: To address the challenges in analytical chemistry instruction, including the complex interconnections among knowledge points, fragmented student learning processes, delayed feedback on learning progress, insufficient personalized support, and the underutilization of course knowledge graphs, a dual-drive teaching model integrating a knowledge graph and an AI teaching assistant was developed. In this model, the knowledge graph presents the course knowledge structure, guides learning pathways, and identifies weak knowledge nodes, while the AI teaching assistant delivers real-time question answering, resource recommendations, and targeted exercises, thereby forming a closed-loop support system of diagnosis, intervention, practice, and feedback. Built on the Chaoxing Learning Platform, a course knowledge graph encompassing 11 chapters, 567 core knowledge points, and 91 teaching resources was constructed, and the AI teaching assistant was trained using over 1,000 course-related question-answer pairs. The instructional practice was evaluated through a descriptive comparison between an experimental class and a control class, supplemented by questionnaire-based feedback. The results showed that the experimental class achieved an average homework score 1.6% higher than that of the control class (P = 0.049 < 0.05), with eight additional students attaining full marks and a 17.2% increase in the full-mark rate (P = 0.037 < 0.05). Among 69 student respondents, 91.3% expressed satisfaction or high satisfaction. This study provides a practical case for the collaborative application of knowledge graphs and AI teaching assistants in the digitally intelligent transformation of analytical chemistry teaching.

Key words: Knowledge graph, AI teaching assistant, Analytical chemistry, Blended learning, Digitally intelligent empowerment