The landscape of physical education has evolved significantly in recent years, reflecting broader cultural shifts toward inclusivity, wellness, and holistic health. Among college courses, basketball has emerged as a popular choice, drawing students with its dynamic energy and emphasis on teamwork. This study aims to apply fuzzy evaluation methods and intelligent design principles to foster innovative development in college basketball courses. Traditional physical education assessment methods often fail to address the diverse needs of students and lack a structured evaluation framework. To address these limitations, this research introduces a novel Refined Pelican-Optimized Fuzzy Neural Network (RPO-FNN) model to efficiently evaluate teaching methods and enhance basketball course delivery. The model uses performance metrics collected from students participating in basketball courses and incorporates teaching specialists’ ratings for each evaluation parameter to provide objective and accurate assessments. By integrating students’ mobility mechanisms and analyzing movement vectors according to functional criteria, the proposed RPO-FNN model effectively evaluates the quality of basketball education. Experimental results demonstrate the model’s superior performance, achieving high precision (98.20%), recall (98.18%), accuracy (98.50%), and F1-score (98.21%), surpassing other existing methods. The study also evaluates metrics such as learning progress, physical fitness, student satisfaction, educational effectiveness, and average skill performance ratings, highlighting the comprehensive capabilities of the RPO-FNN model. Findings indicate that this innovative framework significantly enhances teaching strategies, promotes effective learning, and establishes a strong foundation for advancing sports education practices in college basketball. This approach represents a significant step forward in optimizing physical education through intelligent and data-driven methodologies.
Jun Li (Fri,) studied this question.