面向课堂教学行为的多模态循环分析模型探究  

Research on the Multimodal Cycle Analysis Model for Classroom Teaching Behaviors

作　　者：郑晨虹 张海博 王译鹤 ZHENG Chenhong;ZHANG Haibo;WANG Yihe

机构地区：[1]江西师范大学教师教育高等研究院 [2]浙江广厦建设职业技术大学 [3]贵州师范大学教育学院

出　　处：《吉林省教育学院学报》2025年第3期136-142,共7页Journal of Jilin Provincial Institute of Education

摘　　要：课堂教学行为是课堂教学活动的主体,是影响教学效果的关键要素。在智能教学时代,多模态技术可以实现对课堂教学过程中的言语、动作、生理、环境等行为数据的自动化采集与融合分析,具有客观性、全面性的优势。文章梳理了多模态数据在情感识别、投入测评和互动分析三方面的应用,并构建了多模态课堂教学行为循环分析模型。基于多模态技术的课堂教学行为分析需要利用多模态数据挖掘隐性教学行为,注重多模态数据融合表征教学行为,关注多模态数据使用的安全伦理问题。Classroom teaching behavior is the main subject of classroom teaching activities and the key element affecting teaching effectiveness. In the era of intelligent teaching, multimodal technology can realize the automatic acquisition and fusion analysis of speech, action, physiology, environment and other data in the teaching process, which has the advantages of objectivity and comprehensiveness. The article compares the applications of multimodal data in three aspects: emotion recognition, engagement measurement and interaction analysis, and constructs a multimodal classroom teaching behavior cycle analysis model. The analysis of classroom teaching behavior based on multimodal technology needs to use multimodal data to explore the implicit teaching behavior,focus on the fusion of multimodal data to characterize teaching behavior, and pay attention to the safety and ethical issues of multimodal data use.

关 键 词：课堂教学行为 多模态技术 数据挖掘 智能分析 

分 类 号：G434[文化科学—教育学]