Study on fitness tests of Chinese Northern Jiangsu undergraduate students by conventional and pattern recognition techniques

Abstract

This paper applies pattern recognition and conventional analysis methods to visually and quantitatively study the undergraduate students’ fitness situations in the sampling region of northern Jiangsu of China. We first use the conventional analysis to examine whether the students’ fitness is promoted based on current standards and requirements for sports training by analyzing the changing trends of higher-grades students’ fitness. We also compare the fitness of the same-grade male and females. Secondly, the fitness data are collected under the same standards and operations, so the test data (at various levels) of the same grade should be easy to be separated in the low-dimensional space if both training standards and fitness test operations are effective. We mainly apply the pattern recognition technique to verify this hypothesis through visualizing embedded fitness data to visually perceive the intra- and inter-level distributions. The analysis results illustrate that the higher-grades students’ fitness has the uptrend and the fitness of males is better than that of the same grade females. The visualization on the embeddings of fitness data confirmed the hypothesis and also demonstrated the validity of visualization technique in analyzing the fitness data in sports sciences.