A Prediction Method of Students' Physical Ability Score Based on Improved Interpolation Method

Abstract

The invention discloses a method for predicting students' physical fitness scores with an improved interpolation method. S1 collects the data required for students' physical fitness prediction; S2 uses the MCMC multiple interpolation method to fill in the sparse data; the iteration number N of the MCMC multiple interpolation method is determined The method is as follows: if the convergence of the Markov chain occurs at the 0-kth iteration, the number of iterations when the Markov chain converges is taken as the iteration number N of the MCMC multiple interpolation method; if no Markov chain occurs after the kth iteration The chain converges, and the data D is obtained by the k+i iteration 14 , data D 24 Input the XGBoost model, if the effect of the model is improved, use k+i as the iteration number N of the MCMC multiple interpolation method; S3 uses the data before and after the interpolation to construct a time-segment feature module; S4 is based on the time-segment feature module, using XGBoost to build. model to predict the student's physical fitness score in the second time period. The invention solves the problem of improper calculation of physical fitness prediction data coefficients and interpolation values, and improves the accuracy of test results.

Description

technical field [0001] The invention belongs to the fields of artificial intelligence, data statistics, medical informatization, etc., and relates to a prediction method, a storage medium and a system of a student's physical fitness score with an improved interpolation method. Background technique [0002] A student's physical fitness reflects a student's physical fitness, and students' physical health has always been valued by the state. Schools and parents often pay attention to children's learning and neglect the health of students. At present, many schools will test the physical fitness of students every year to understand the physical fitness status of students. In the prior art, the following methods are adopted in the physical fitness test field for students: in projects such as lung capacity, 50-meter running, sitting body forward bending, one-minute rope skipping, turning back and running, sit-ups, step test, pull-ups, standing long jump and ball games Among them,...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the embodiments of the present invention is to solve the technical problems that the existing student physical fitness detection method is time-consuming and labor-intensive, unable to effectively process data with a large number of dimensions, large density differences, and a high proportion of missing values, and thus it is difficult to predict the physical state of students in time , providing a method for predicting students' physical fitness scores by improving the interpolation method, comprising the following steps:

Images