Journal of Higher Education Research

With the increasingly prominent individualized differences in physical health problems among college students, the traditional "one size fits all" physical education teaching model is unable to meet the needs of precise intervention. This study developed a big data-driven precision teaching model for college students' physical fitness and health. Based on multi-source data such as physical fitness testing, campus running systems, and wearable devices, a five-level closed-loop system was established, consisting of a data layer, analysis layer, decision-making layer, implementation layer, and feedback layer. This model divides students' physical types (such as weak endurance/insufficient strength) through clustering algorithms, intelligently generates personalized exercise prescriptions based on weak indicators, and designs a linkage mechanism and dynamic adjustment rules between in class and out of class activities. Taking overweight students with insufficient endurance as an example, the effectiveness of the mapping rule library between physical fitness indicators and exercise parameters was verified. The model has the advantages of low technological threshold, teacher empowerment, and precise intervention, providing a new path to solve the contradiction of "weak physical fitness cannot train and strong physical fitness cannot eat enough", and providing technical support for the implementation of sports education integration policies.