Research on Student Learning Behavior in Smart Classroom Based on Multimodal Data
The smart classroom empowered by emerging technologies provides rich educational multimodal data for student learning behavior monitoring,evaluation,feedback,and early warning.Fully collecting,processing,and analyzing educational multimodal data can provide reference for research and practice related to education and teaching.Firstly,a multimodal data analysis framework based on the three-dimensional dimensions of sound data,image data,and text data is constructed by reviewing existing research on intelligent classroom learning behavior analysis and multimodal data teaching,and drawing on the classification system of learning behavior.The framework presents changes in student learning behavior in intelligent classrooms from four aspects,namely speech learning activities,position movement,body movements,and technology use;Secondly,the multimodal data reflecting the characteristics of student learning behavior will be encoded and qualitatively represented to form an encoding system;Finally,from the perspective of situational and temporal behavior analysis,the frequency and periodic changes of multimodal data on learning behavior in lesson examples is analyzed.Research has shown that in a smart classroom environment,student participation,initiative,and focus in classroom learning are constantly improving,and the classroom learning atmosphere is developing in a positive direction,with ineffective and irrelevant learning behaviors gradually decreasing.In addition,by analyzing multimodal data on student learning behavior,it is aimed to provide data sources and scientific basis for students to understand their own learning behavior and status,personalized teaching for teachers,and the construction of learning evaluation systems for researchers.
NETL
NSTL
万方数据
维普
