Research on the optimization system of athlete selection and training effect based on big data

Accurate talent selection and scientific training arrangements are a very important part of the athlete training process. The data generated by athlete selection and training has long been underutilized, so this paper designs a K-means clustering algorithm based on optimizing the initial clustering center and profile coefficients, to cluster and analyze the performance of athlete selection indexes. A collaborative filtering algorithm and a content-based recommendation algorithm are also combined to recommend suitable training programs for athletes to help them develop themselves. On this basis, an athlete selection and training optimization system is designed to improve the effect of athlete development. Five athlete categories are obtained by clustering according to the improved clustering algorithm, and effective evaluation of the performance of different athlete groups in the selection test is achieved. The personalized training algorithm designed in this paper achieves the lowest RMSE value, the recommendation effect is more accurate, and 7 out of 8 recommendation results match the actual situation, which helps to improve the athlete selection and training settings.