基于多模态数据融合的课堂教学行为分析研究

Research on the Analysis of Classroom Teaching Behaviors Based on Multimodal Data Fusion

课堂教学行为是影响课堂教学效果的重要因素,然而现有研究在分析课堂教学行为方面存在一些问题,如数据类型单一、编码复杂以及难以发现课堂运转规律等。人工智能技术为课堂教学行为数据的大规模采集和处理提供了全新的机会。通过运用人工智能技术对某省的10所学校917个智慧课堂教学视频进行了分析,并揭示了课堂教学行为的一些特征。研究结果表明:课堂教学行为的频率和类型存在较大差异,不同教学行为之间存在相关性,同时教师行为和学生行为也不完全独立。这些研究结果为人工智能时代的课堂教学规律挖掘、课堂教学改进以及教研活动的开展提供了重要的参考依据。

The classroom teaching behavior is an important factor affecting classroom teaching effects.However,existing research has some problems in the analysis of classroom teaching behaviors,such as the single data type,complex coding and the difficulty in discovering classroom operation rules.Artificial intelligence technology provides a new opportunity for the large-scale collection and processing of classroom teaching behavior data.In this study,artificial intelligence technology is used to analyze 917 smart classroom teaching videos from 10 schools in a certain province,revealing some characteristics of classroom teaching behaviors.Research results show that there are great differences in the frequency and types of classroom teaching behaviors,there is correlation among different teaching behaviors,and that teacher behaviors and student behaviors are not completely independent.These research results provide important reference for the mining of classroom teaching rules,the improvement of classroom teaching,and the development of teaching and research activities in the era of artificial intelligence.