A Study on Classification and Instructional Adjustment of Students' Physical Fitness Test Data Based on Cluster Analysis

Abstract

Enhancing students' physical fitness level is the current focus of physical education teaching, and how to do a good job of teaching physical fitness training is also a problem that physical education teachers must think about. In this paper, the physical fitness test data of undergraduates of grade 2023-2024 of University A in 2025 is selected as an example, and the physical fitness test data is divided into male and female groups, and the improved K-means method is applied to cluster analysis of the two groups of data, and outlier analysis of the clustering results is carried out. Then the buddy recommendation model based on set-pair theory was proposed to get the degree of recommendation between users, and the buddy recommendation model based on set-pair theory was utilized to design the students' physical test teaching intervention experiment, and the data of various indexes before and after the experiment were statistically analyzed. The results of the study show that there are differences between male and female groups in the physical fitness test, and the percentage of outlier students whose physical fitness data is different from the majority is found to be about 1%, and the buddy recommendation model based on set-pair theory has a better holistic effect and stability than the traditional recommendation algorithms in the recommendation of exercise prescription. The experimental conclusion verifies that the exercise prescription generation and intervention monitoring system has a certain role and scientific validity in promoting students' health through exercise intervention.