【摘要】 新工科教学要求以学生为中心，并可以从多个维度对学生的学习成果进行评价。结合在线教学工具，文章提出了基于数据挖掘方法的学生学习轨迹分析框架。通过记录教学过程中学生的学习活动，基于Adaboost方法构建学生学习成绩预测模型，并对学生的学习进展进行分析和评估。在此基础上针对性地采用客户需求分析、场景化教学升级、多元化角色扮演等教学干预措施。结果表明所构建的模型能够对大多数学生的成绩进行准确预测，并且所采用的干预措施能够有效提升学生成绩和教学质量。更多还原

【Abstract】 The emerging engineering education emphasizes student-centred approaches and the ability to evaluate student’s learning outcomes from multiple dimensions. Combined with online teaching tools, a framework of student learning trajectory analysis based on data mining method is proposed. By recording student’s learning activities during the teaching process, the prediction model of students’ academic performances is constructed based on the Adaboost method and the student’s learning progress is analyzed and evaluated. Based on this, targeted teaching intervention measures such as customer demand analysis, scenario-based teaching and diversified role-playing are adopted. The results show that the constructed model can accurately predict the majority of student achievements and the intervention measures employed can effectively enhance student performance and teaching quality.更多还原