Recently, the 2024 Guangdong-Hong Kong-Macao Greater Bay Area Artificial Intelligence Industry Conference, guided by the Guangdong Provincial Department of Science and Technology and hosted by the Guangdong Artificial Intelligence Industry Association, was successfully held in Guangzhou. The conference, themed "Building the Foundation with General AI, Empowering Thousands of Industries to Set Sail," attracted numerous experts, scholars, and industry representatives to discuss the latest developments and application prospects of the artificial intelligence industry.

During the conference, the 2024 Guangdong Artificial Intelligence Industry Association Science and Technology Awards ceremony was grandly held. The "Key Technologies and Applications of Online Classroom Teaching Quality Evaluation Based on Multimodal Machine Learning" (hereinafter referred to as "the technology"), jointly developed by TAL Education Group and Jinan University, won the First Prize of the Science and Technology Progress Award.

The technology designs multimodal data representation layers based on the audio, video, text, and image data from the teaching process, enabling effective collection and labeling of teaching behaviors in real educational scenarios. It addresses low-quality multimodal data issues in the education field, such as audio and video desynchronization, modal misalignment, and data loss. The technology improves prediction accuracy on a given dataset by 10% compared to the optimal algorithm at that time. Ultimately, it integrates several advanced AI+education capabilities, including automatic recognition of in-class teaching behaviors, automatic classroom teaching quality rating, and automatic alert systems for problematic classrooms. Additionally, it offers applications like automatic generation of student learning reports, teacher lesson preparation evaluation, and student oral ability assessment, successfully breaking the "black box" of the classroom teaching process and making classroom evaluation indicators more objective and quantifiable.

Since the project's implementation, it has received high praise and recognition from various parties. Schools and institutions can leverage the technology and applications for online classroom teaching quality evaluation to achieve large-scale, high-quality classroom inspections, ensuring classroom quality and compliance. Teachers can use it to adjust their teaching content and methods, thereby guaranteeing the classroom experience and quality while improving their teaching abilities, contributing to the development of a high-quality teaching team. Parents and students can use the system-generated post-class deep learning reports to better understand classroom learning conditions and students' shortcomings, facilitating adjustments for personalized learning and steady growth. Students can also use the system's oral assessment capabilities to improve their expressive skills, better communicating their needs, which in turn helps parents and teachers understand them better, enhancing teacher-student emotional bonds and improving parent-child relationships.

The project has also received recognition and affirmation from both domestic and international government agencies. The technology solution was selected as a representative case of China's exploration in the AI+education field in the UNICEF report, "Artificial Intelligence for Children." It has been recognized by the Ministry of Education, the Ministry of Science and Technology, and the Office of the  Central Cyberspace Affairs Commission, included as a representative application case in the "2021 AI Enterprises Typical Application Case Study" by the Office of the  Central Cyberspace Affairs Commission, and highlighted in the Ministry of Science and Technology's "Intelligent Education Innovation Application Development Report." It was also showcased as an achievement of the National New Generation AI Open Platform for Smart Education at the "the Thirteenth Five-Year Plan" Science and Technology Innovation Achievements Exhibition. Moreover, it has received acknowledgments from numerous social organizations, being included among the excellent AI achievements at the BAAI 2020 Beijing Zhiyuan Conference and featured in the "Smart Education Development and Industry Landscape Research Report (2021)" by the China Internet Association, as well as in the UNICEF's "AI for Children" excellent cases collection.

The project's technical capabilities and research outcomes have also received international recognition, winning accolades such as the Best Resource Paper Nomination Award at the 2022 ACM International Conference on Information and Knowledge Management (CIKM), first place in the 2022 International Conference on Educational Data Mining (EDM) Student Performance Prediction Challenge, second place in the Knowledge Tracing Challenge, and first place in the 2023 International AI in Education Competition "The Quest for Quality Questions."

Currently, the project's technical capabilities are available through the National Open Innovation Platform for Next Generation Artificial Intelligence, built by TAL Education Group.

The link of Jinan University：https://news.jnu.edu.cn/content/202411/27/c3820.html