AI-driven innovation in the teaching model of Microbiology Experiment and construction of evaluation systems

Abstract:

Against the backdrop of the in-depth integration of educational information and intelligent technologies, conventional teaching models of Microbiology Experiment face challenges such as outdated teaching contents, lagging dynamic feedback, and single-dimensional evaluation criteria. In this study, we proposed an artificial intelligence (AI)-driven innovation framework to design personalized learning pathways and establish teaching plans based on foundational experiments and incorporating interdisciplinary collaboration, practical application, and innovative thinking cultivation. An intelligent diagnostic platform of experimental data was developed to precisely identify operational errors, data processing flaws, and analytical inaccuracy across experimental stages, enabling a closed-loop teaching mechanism of diagnosis-intervention-feedback. Furthermore, we constructed a dynamic evaluation system with a three-tier evaluation index system encompassing basic technical skills, innovative thinking, and scientific literacy, supported by multidimensional, visualized, and continuous evaluation methodologies. Practices have proved that this AI-participated teaching model significantly improves the course grades, teaching efficiency, and participation. The teaching reform effectively bridges the gaps between conventional teaching and practical needs, being of great significance for cultivating microbiology talents with innovative thinking and practical competency.