Analysis of Classroom Teaching Behaviors Based on Multimodal Data Model

DOI: https://doi.org/10.62517/jike.202404407

Author(s)

Xinrui Ma, Jingxia Chen*

Affiliation(s)

Beijing Union University, Beijing, China, *Corresponding Author

Abstract

Precise identification of classroom behavior can help teachers and students understand classrooms, and help promote the development of smart classrooms. This article designed a multi -mode data model supported by multimodal data support based on classroom teaching scenes, including classroom teaching layers, data collection layers, algorithm analysis layers, and application service layers. Analysis of classroom teaching behaviors, this article extracts the image characteristics in the video based on the deep learning algorithm of YOLO-V5, the voice recognition technology extracts voice characteristics. Multi -mode data model analysis has obtained good analysis results and recognition results. In order to verify the effectiveness of the selected model, the model performance was performed on the labeled classroom behavior data set. The test results show that the selected model shows good performance in the analysis and identification of classroom behavior in the education scene.

Keywords

Multi-Mode Data; Analysis of Classroom Teaching Behavior; Behavior Recognition; Smart Classroom

References

[1] Flanders N A. Intent, action and feedback: A preparation for teaching. Journal of Teacher Education,1963,3(14):25-260. [2] Xiaoqing Gu, Wei Wang. New exploration of classroom analysis technology to support teachers' professional development. China Electronic Education, 2004(7): 18-21. [3] Jianfeng Jin, Xiaoqing Gu. Analysis and Research on Classroom Teaching Behavior in the Information Technology Environment. China Electronic Education, 2010(9): 82-86. [4] Jing Li, Qi Zhang, Zhigang Miao, etc. Research on Interactive Behavior of Information Classroom Teaching in Middle Schools: A Qualitative Analysis Perspective. China Electronic Education, 2014(2): 101-107. [5] Su Mu, Pingping Zuo. Research on the Analysis Method of Classroom Teaching Behavior in the Information Technology Teaching Environment. Research on Electronic Education, 2015, 36(9): 62. [6] Zhong Sun, Kaiyue Lv, Luo Liming, etc. Analysis of Classroom Teaching Based on Artificial Intelligence. China Electronic Education, 2020(10): 15-23. [7] Xinyun Liu, Shiping Ye, Denghui Zhang. Improved multi-objective regression method for detecting student classroom behavior. Computer Engineering and Design, 2020. [8] Guoqing Lu, Kui Xie, Qingtang Liu, etc. Analysis of Classroom Teaching Behavior Based on Artificial Intelligence Engine Automatic Annotation. Open Education Research, 2021. [9] Prieto L P, Sharma K, Dillenbourg P, et al. Teaching analytics: towards automatic extraction of orchestration graphs using wearable sensors. Proceedings of the sixth international conference on learning analytics & knowledge, 2016: 148-157. [10] Riquelme F, Munoz R, Mac Lean R, et al. Using multimodal learning analytics to study collaboration on discussion groups: A social network approach. Universal Access in the Information Society, 2019, 18: 633-643. [11] Lele Zhang, Xiaoqing Gu. Classroom teaching behavior analysis model and practical framework supported by multimodal data. Open Education Research, 2022, 28(06): 101-110.13966/j.cnki.kfjyyj.2022.06.011. [12] Yashu Chen. Multidimensional Classroom Interaction Analysis Based on Speech Recognition. Huazhong Normal University, 2021. [13] Zhihui Zhao, Yi Zhou, Weihong Li, Chaohui Tang, Qiang Guo, Rigao Chen. Construction of TCM syndrome differentiation model of type 2 diabetes based on deep learning multimodal fusion. World Science and Technology - Modernization of Traditional Chinese Medicine, 2024, 1-1.