| 刊名 | Meteorological and Environmental Research |
| 作者 | Wei HUANG, Zhaoliang LIU, Binjun GAN, Na ZHAO, Daobo WANG, Guoren LAO* |
| 作者单位 | College of Smart Agriculture, Yulin Normal University/Key Laboratory of Smart Agriculture for Guangxi Characteristic Horticultural Crops/Guangxi Zhuang Autonomous Region Facility Agriculture Engineering Research Center, Yulin 535000, China |
| DOI | 10.19547/j.issn2152-3940.2026.02-03.19 |
| 年份 | 2026 |
| 刊期 | 3 |
| 页码 | 86-89 |
| 关键词 | AI; Agricultural master; Affective computing; Practical classroom engagement; Multimodal data |
| 摘要 | The evaluation of agricultural master's practice teaching has long suffered from issues such as a heavy focus on outcomes over processes, single-dimensional criteria, and strong subjectivity, making it difficult to effectively measure students' classroom engagement and innovation capabilities. Addressing the limitations of traditional evaluation models, by introducing AI affective computing technology into the field of agricultural master's practice teaching evaluation, a comprehensive evaluation system based on multi-dimensional affective computing is constructed. In this paper, the evaluation index system is deconstructed from three core dimensions: focus, collaboration, and innovative behavior, establishing a mapping relationship between affective computing technology and evaluation dimensions. Building on this, a multimodal data collection scheme is designed, employing deep learning algorithms to extract key features of engagement and innovative behavior, thereby constructing a comprehensive evaluation model. Finally, implementation pathways for the evaluation system are proposed. It demonstrates that this system can dynamically track and accurately characterize students' practical processes, providing scientific and data-driven support for improving the quality of agricultural master's talent cultivation. |
