A Fuzzy Neural Network-Based Evaluation Method for Physical Education Teaching Management in Colleges

Area of Science:

• Educational Technology
• Artificial Intelligence in Education
• Sports Science

Background:

• Evaluating physical education (PE) teaching quality in universities requires accurate and valid methods.
• Existing assessment models may lack the precision needed for nuanced evaluation of PE instruction.
• The integration of advanced computational models can potentially enhance educational assessment.

Purpose of the Study:

• To introduce a novel Fuzzy Neural Network (FNN) model for assessing the teaching quality of physical education (PE) in colleges and universities.
• To improve the validity and accuracy of PE teaching quality evaluations.
• To compare the performance of the FNN model against traditional assessment methods.

Main Methods:

• Development of a multi-index assessment process using the analytic hierarchy process (AHP) for university PE teacher performance, considering teaching material, method, attitude, and effect.
• Implementation of a Fuzzy Neural Network (FNN) model utilizing score data as parameters to classify teaching quality into categories (excellent, good, average, low).
• Comparative analysis of the FNN model's classification accuracy, specificity, sensitivity, and F1 score against other assessment methods.

Main Results:

• The proposed FNN model achieved high performance metrics: 96% accuracy, 95% specificity, 90% sensitivity, and a 94% F1 score.
• FNN demonstrated superior classification performance compared to other methods in evaluating PE instructional excellence.
• The study confirmed the effectiveness of the proposed approach through comparison with standard PE teaching strategies.

Conclusions:

• The novel Fuzzy Neural Network (FNN) model offers a significant advancement in the objective assessment of physical education teaching quality in higher education.
• The FNN model provides a more accurate, valid, and effective tool for evaluating PE instructors' performance.
• This AI-driven approach holds promise for enhancing educational assessment practices in physical education and potentially other disciplines.