A theoretical and practical study on the components of intelligent physical education courses in colleges and universities based on Bayesian non-parametric model

Abstract

This paper performs feature extraction based on the Bayesian nonparametric model for smart sports courses to analyze the Gaussian process principle for composing random variables. The probability distribution of the parameters is estimated based on the known data set, given the input data vector to decompose the likelihood function. To calculate the posterior probabilities of the parameters, the kernel function features are analyzed to give the expressions of eigenvalues and eigenvectors, and data normalization is performed. According to the results, the experimental group had higher sports test scores than the control group, and the third test score of high-speed running reached 9 points. Therefore, colleges and universities should focus on cultivating students' innovative thinking abilities and helping them master sports knowledge and skills to achieve their educational goals.

Keywords

• bayesian nonparametric
• random variables
• gaussian distribution
• likelihood function
• smart physical education curriculum
• 97q70