ORIGINAL RESEARCH article
Front. Artif. Intell.
Sec. Machine Learning and Artificial Intelligence
Volume 8 - 2025 | doi: 10.3389/frai.2025.1553051
An AI-Powered Framework for Assessing Teacher Performance in Classroom Interactions: A Deep Learning Approach
Provisionally accepted
- 1Faculty of Computing and Information Technology, Department of Computer Sciences, King Abdulaziz University, Jeddah, Saudi Arabia
- 2Ministry of Education, Riyadh, Saudi Arabia
- 3Immersive virtual reality research group, King Abdulaziz University, Jeddah, Saudi Arabia
- 4Center of Research Excellence in Artificial Intelligence and Data Science, King Abdulaziz University, Jeddah, Saudi Arabia
Select one of your emails
You have multiple emails registered with Frontiers:
Notify me on publication
Please enter your email address:
If you already have an account, please login
You don't have a Frontiers account ? You can register here
Teacher performance evaluation is crucial for enhancing teaching quality and supporting professional development. Traditional methods often suffer from subjectivity, time-consuming processes, and limited reliability. In this paper, we present an AI-powered Framework for Assessing Teacher Performance in Classroom Interactions using deep learning techniques. Our framework employs three state-of-the-art object detection algorithms: YOLOv8, Faster R-CNN, and RetinaNet, to detect and analyze eleven in-classroom interactions. A labeled dataset of 7259 images collected from actual classrooms was used to train and evaluate the models. Results indicate that YOLOv8 models outperform others, achieving a mAP of 85.8%, demonstrating their effectiveness in detecting various classroom interactions with high accuracy. This study highlights the potential of AI-based systems in providing objective and reliable assessments of teacher performance, thereby contributing to improved educational outcomes.
Keywords: data curation, A.Almubarak., W.Alhalabi. and E.Alharbi., formal analysis, I.Albidewi. and W.Alhalabi., funding acquisition, W.Alhalabi., investigation, A.Almubarak. and E.Alharbi., methodology, A.Almubarak
Received: 30 Dec 2024; Accepted: 04 Aug 2025.
Copyright: © 2025 Almubarak, Alhalabi, Albidewi and Alharbi. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
* Correspondence:
Arwa Almubarak, Faculty of Computing and Information Technology, Department of Computer Sciences, King Abdulaziz University, Jeddah, Saudi Arabia
Wadee Alhalabi, Faculty of Computing and Information Technology, Department of Computer Sciences, King Abdulaziz University, Jeddah, Saudi Arabia
Eaman Alharbi, Faculty of Computing and Information Technology, Department of Computer Sciences, King Abdulaziz University, Jeddah, Saudi Arabia
Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.