职业教育数据驱动的精准教学与评价

Data-Driven Precise Teaching and Evaluation in Vocational Education

摘要: 职业教育数字化转型的核心在于利用数据智能实现规模化因材施教。针对当前精准教学实践中存在的数据孤岛、分析表层化、干预滞后、评价单一等问题，本文提出一个名为智慧教学决策支持系统的集成化技术框架。该框架包含多源数据感知与融合层、智能分析中枢层、精准干预与动态评价层。其中，智能分析中枢层创新性地整合了基于图神经网络的学业风险预测模型与过渡网络分析，实现了对学生多维表现及学习过程时序模式的深度挖掘。本文以高职《Python程序设计》课程为案例，展示了该框架的实施全流程。实证结果表明，应用该框架的班级在通过率、学习投入度及高维能力发展上均显著优于对照组。本研究为构建教学、评价、干预一体化提供了可落地的技术路径与实证参考。

Abstract: The core of the digital transformation of education lies in leveraging data intelligence to realize large-scale personalized teaching. Aiming at the prevalent problems in current precision teaching practices, including data silos, superficial data analysis, delayed intervention and simplistic evaluation, this paper proposes an integrated technical framework named the Intelligent Teaching Decision Support System. The framework consists of three layers: multi-source data perception and fusion, intelligent analysis hub, and precise intervention and dynamic evaluation. Within the intelligent analysis hub, a graph neural network-based academic risk prediction model and transition network analysis are innovatively integrated to conduct in-depth mining of students’ multi-dimensional performance and temporal patterns of learning processes. Taking the college course Python Programming as a case study, this paper elaborates on the entire implementation process of the framework. Empirical results demonstrate that the classes adopting this framework achieve significantly better outcomes than the control group in terms of pass rates, learning engagement and the development of high-dimensional competencies. This study provides a practical technical pathway and empirical reference for constructing an integrated system of teaching, evaluation and intervention.