智慧课堂数据挖掘分析与应用实证研究

Empirical Study on Analysis and Application of Smart Classroom Data Mining

文章基于智慧课堂大数据,利用教育数据挖掘技术进行建模、分析和处理,从教学设计过程、学习活动过程和学习结果三个层面进行整体设计,包括基于学习者行为建模与分析的"1+3模式"以及基于学习内容和结果建模与分析的"3+1模式",构建了学习行为影响分析、学习行为路径分析、学生行为关联性分析和学业成绩预测分析四类应用模型,并基于真实数据对智慧课堂数据挖掘应用进行实证分析。以学生的行为路径分析为例,得出成绩上升的学生行为共同模式是先学习微课再收藏,而成绩下降的学生则相反,收藏并不一定会学习,而学习行为发生在收藏前则表明成绩上升的学生肯定了微课对自己的价值,这也从一定程度上解释了学业成绩变动的原因。

Based on big data of smart classroom, this article uses educational data mining technology to model, analyze, process and makes integrated design from three aspects including the process of instructional design, learning activities and learning results. Specifically, the modeling and analysis of ＂1＋3 model＂ based on learner behavior and that of ＂3 ＋1 model＂ based on learning content and result are described. Then this paper constructs four application models, namely the impact analysis of learning behavior, the path analysis of learning behavior, the correlation analysis of student behavior and the prediction analysis of academic achievement and makes an empirical analysis on the application of smart classroom data mining based on real data. Finally, take the path analysis of student＇s behavior as an example, this paper concludes that the common model of behaviors for students with high achievement is to learn micro lessons first and then save it. While for students with lower grades, they would just save the course. That the learning behaviors of students with high achievement happen before they save those micro lessons indicate that they have believed the value of those lessons, which, to a certain extent, explains the reasons for the changes in their academic performance.