An empirical study on the improvement of students’ physical fitness and health by college physical education programs based on the background of big data

The reform of college physical education courses in the context of big data is of great significance to improve the quality of teaching and meet the needs of students. The study is based on the K-medus clustering algorithm to personalize the teaching content of college physical education courses. The standard deviation is used to define the initial centroid candidate set, and the initial centroids are determined in a stepwise increasing manner, which ensures that the sample points with greater densities are selected as the initial clustering centroids. Students with similar body types are clustered together by the method, and teachers can create targeted individualized teaching content based on students with different body types. After the implementation of personalized teaching, the physical fitness of both boys and girls improved. The excellent and good rates of boys’ physical health increased by 7.75% and 4.34%, respectively. The excellent and good rates of physical health among female students increased by 14.03%. It shows that students’ physical fitness has significantly improved after reforming the physical education program in the context of big data.