Research and Analysis of Physical Health Test for Students Based on C4.5 Algorithm in Universities with Industry Characteristics

In this paper, the hierarchical sensitivity analysis is constructed for a college student's physical health test dataset using the C4.5 decision tree algorithm. The dataset is from three consecutive years in 2017 and 2018 grades at an industry characteristics university. The analysis results indicate that the grade and gender factors play a decisive role in the dataset's attributes. On this basis, the pruning process is adopted for the algorithm. Pearson correlation coefficients show that industry characteristics majors are positively associated with vital capacity and long-distance running during freshman and sophomore years. Besides, the Spearman correlation coefficient analysis results show that industry characteristics majors positively correlate with vital capacity, 50-meter running, long-distance running, and overall performance, which has a positive effect. The analysis can provide a scientific basis for the physical health education of college students in universities with industry characteristics.