Data-Driven Reconstruction of Classroom Teaching Evaluation in Higher Education: A Framework Integrating Multimodal Analytics and AI

In the digital age marked by advancements in AI, big data, and IoT, higher education quality evaluation is undergoing transformative changes. Drawing on teaching evaluation data from 32 universities, this study proposes a three-tier digital intelligence evaluation model integrating intelligent perception, data analysis, and decision support. The model deeply incorporates technologies like facial recognition, learning analysis, and knowledge graphs into educational assessment, adhering to principles of data-driven decision-making, quality focus, and multi-stakeholder collaboration. It reconstructs evaluation frameworks in content, methodology, and technology through innovative data integration methods, optimized indicator systems, and advanced analytical tools. Key strategies include digital transformation of evaluation processes, refinement of performance metrics, and enhanced data governance. The system not only improves evaluation efficiency and precision but also supports teacher professional development while offering insights for quality improvement. As technology evolves, continuous optimization of this digital intelligence framework will drive higher education toward smarter, more personalized, and sustainable development.