Application of Classification Algorithm Based on Naive Bayes in Data Analysis of Fitness Test

Abstract

Abstract An important aspect of physique data analysis is to provide a basis for developing national sports and improving physical education. This article aims to explore the application of Naive Bayesian classification algorithm in physical test data analysis, in order to conduct differentiated teaching and training from person to person, so as to effectively improve the physical health of students. This article uses the real physical test data of our school in 2018 and 2019 to form a complete data set after coding to construct a student physical classifier. In this way, the correct judgment of the student’s physical condition can be realized with a certain probability, and the problems of the student’s physical health can be detected and warned in time. Using the slicing method to cut out 20% as the test data, the classifier is tested. From the test results, it can be seen that the comprehensive correct rate of the physique classifier based on the naive Bayes algorithm reaches 81.02%. The research in this article provides help to better promote the physical fitness testing of colleges and universities, and provides a new way of thinking and practical methods for improving the quality of physical education teaching.