Prediction of College Students' Physical Test Effect Based on BP Neural Network
Min Xiao, ZeYu Li, Zheng Sun, Liu Xunzi · 2022 5th International Conference on Data Science and Information Technology (DSIT) · 2022
With economic advancements in our country, people's aspiration for a better life fuels their enthusiasm for sports, and thus physical exercise has become increasingly important in people's minds. Meanwhile, China is also promoting the physical fitness test of college students nationwide, with the view to encouraging regular physical exercise among college students. According to the survey, the 800/1000-meter long-distance running has become the most challenging test for college students. As long-distance running is physically demanding, if students seldom take on exercises, some of them are likely to experience physical discomfort during the test. To raise college students' awareness of physical fitness and set goals for their daily exercise, this paper studied the impact of students' gender, height, weight, and vital capacity, etc. on their performance of the 800/1000-meter running. From the students who passed the physical fitness test of Three Gorges University in 2016, 60 groups of test results of these students were randomly selected as the research object. By analyzing the connections between these data, a prediction model was built. Subsequently, through K-Fold cross-validation, the students' actual scores of 800/1000-meter running and predicted values were compared and analyzed. The experimental results showed that the BP neural network prediction model can effectively predict the performance of students in the 800/1000-meter long-distance running. The predicted results can provide a reference for college students' to set goals for daily exercise, thus arousing their interest in daily exercise.