基于数据挖掘的校园运动研究

Research on Campus Sports Based on Data Mining

体质测试作为反馈大学生体质健康水平的根本途径,为高校开展学生健康干预工作提供了数据支撑,但如何对体测数据进行科学分析及合理使用也变得尤为重要。文章通过数据挖掘技术研究大学生体测数据,分别采用决策树、朴素贝叶斯、贝叶斯神经网络对体测数据进行预测,结果显示,贝叶斯神经网络的预测准确率最高。利用CART决策树对体测数据进行分类,由此可得到最优决策树,由最优决策树分析影响大学生体质水平的重要因素,进一步探讨体测成绩对大学生身体素质的影响和作用,从而提高大学生参与校园运动的热情和兴趣。

Physical fitness testing,as the fundamental way to provide feedback on the physical health level of college students,provides data support for universities to carry out student health intervention work.However,it has become particularly important to scientifically analyze and reasonably use physical fitness data.This paper uses data mining techniques to study the physical measurement data of college students,and uses decision trees,naive Bayes,and Bayesian neural networks to predict the physical measurement data.The results show that Bayesian neural networks have the highest prediction accuracy.By using the CART decision tree to classify physical testing data,the optimal decision tree can be obtained.It analyzes the important factors that affect the physical fitness level of college students through the optimal decision tree,further explore the impact and role of physical testing scores on the physical fitness of college students,and thereby enhance their enthusiasm and interest in participating in campus sports.