Artificial intelligent based video analysis on the teaching interaction patterns in classroom environment

Abstract

Recently, the development of technology has enriched the form of classroom interaction. Exploring the characteristics of current classroom teaching interaction forms can clarify the deficiencies of teaching interactions, thereby improving teaching. Based on the existing classroom teaching interactive coding system, this paper adopted ITIAS coding system, and took classroom with interactive whiteboard, interactive television or mobile terminals as research scene, selected 20 classroom videos of teaching cases in this environment as research objects. Computer vision, one of the artificial intelligent technologies was applied for video analysis from four aspects: The classroom teaching atmosphere, the teacher-student interaction, the student-student interaction, the interaction between human and technology. Through cluster analysis, three clusters of sample’s behavioral sequences were found. According to the analysis on the behavioral sequences and the behavioral transition diagram of each cluster, three classroom teaching interaction patterns were identified, including immediate interaction pattern, waiting interaction pattern and shallow interaction pattern.