Evaluation of higher education teaching effectiveness based on fuzzy algorithms and artificial intelligence

Subjective biases and unreliable evaluation procedures make evaluating higher education faculty challenging. A Fuzzy-AI-Metaheuristic Framework for Teaching Assessment (FAM-Teach) is presented in this research to assess instructional performance adaptively, data-driven, and objectively. It uses AI to analyze massive amounts of feedback data, fuzzy logic to manage qualitative evaluation uncertainty, and the Firefly Algorithm (FA) to improve indicator interaction and weighting. Instruction, student involvement, fair evaluation, and measurable learning outcomes are essential for teaching success. FA optimisation’s quicker convergence and precise parameter weight calculation allow the model to adapt to changing instructional conditions. Experimental validation with 1,200 students from five departments showed that FAM-Teach outperforms traditional methods in performance, consistency, and interpretability. The data were preprocessed, and incomplete data were removed, resulting in 650 instances for training and testing. Success requires participants, timely feedback, and tailored training, according to the study. The findings demonstrate that FAM-Teach is a smart and scalable technique to assess higher education teaching using evidence.