Physical Fitness Clustering Analysis Based on Self-Organizing Feature Maps Network

Abstract

Clustering analysis based on self-organizing feature maps (SOM) network has been widely used in various areas of cluster analysis. In this paper, this network is applied to the clustering analysis of students' physical level. The software is used to study and train the designed self-organizing feature maps network. Correspondingly, Neural Network Model, and the physical measurement level of three levels of classification (The first level is good, the second level is qualified, the third level is unqualified), to achieve the level of physical cluster analysis. The results show that the self-organizing feature maps network can automatically classify the physical test scores unsupervised learning, and visually and clearly see the level classification of the physical test scores, analyze the main factors affecting physical fitness from the clustering analysis of physical test results.