AIGC-Driven Teaching Quality: A Closed-Loop Feedback Framework Integrating RFM Stratification and Education Prompt Engineering

This study constructs an AIGC-based teaching quality evaluation framework to overcome traditional limitations and provide a scientific basis for teaching improvement. It uses a closed-loop feedback mechanism of “data-model-strategy”. It integrates structured prompt engineering to transform teaching data (including student evaluations, RFM stratification, and student comments) into actionable improvement strategies. The process includes “dynamic stratification (RFM) + semantic parsing (NLP) + strategy generation (LLM)”, with a three-round progressive dialogue model (RTF-SCOPE-T) for precise teaching diagnosis. Conducted on the “Corporate Behavior Simulation” course at University G, the quasi-experimental design with experimental and control groups for comparative analysis. Results show that teachers in the experimental group, guided by AIGC-generated teaching quality evaluation reports, significantly improved their teaching evaluation scores, while the control group exhibited no significant changes. AIGC technology, through multi-dimensional data collection and analysis, significantly enhanced the reliability and objectivity of the evaluation reports, aiding teachers in accurately identifying issues in their teaching and facilitating effective reflection.