基于生理指标监测的U19、U17足球运动员训练效果干预研究——机器学习判别应用

Based on the Monitoring of Physiological Parameters U19, U17 Football Player Training Effect Intervention——Machine Learning Judgment Application

通过机器学习判别方法评估生理指标监测过程对U19、U17梯队足球运动员训练效果产生的干预作用及干预程度。结果表明:实施生理监测干预的运动员梯队与未实施监测梯队在技术统计数据上表现出多元显著差异(u19T=21.56>??28X 0.05、u17T=29.37>??28X 0.05)。随着生理监测统计指标的增加,SVM机器学习算法对干预过程的判别正确率表现出提高趋势。增加Trimp、O2max指标,U19、U17梯队预测判别正确率分别上升12.5%、25%;增加HB、CK、T指标,U19梯队判别准确率上升25%,U17梯队判别准确率保持不变;增加HR*VO2max交互指标,U19梯队判定正确率达87.5%,U17梯队判别准确率达100%。结论:引入生理指标监控的足球训练过程可高效反馈机体的应激反应,为运动训练提供辅助支持。

The evaluation of the U19, U17 tier footballer intervention training effect generated and the degree of intervention physiological indicators for monitoring the process by machine learning identification method. The results showed that： physiological monitoring the implementation of the intervention echelon athletes and non athletes echelon in monitoring the implementation of the technical multivariate statistics showed significant differences （ Tu19〉X8^2 （ 0.05 ）, Tul7〉 X8^2 （0.05）） With the increase in physiologic monitoring statistical indicators, SVM machine learning algorithms to determine the correct rate forecasting group showed a trend of increasing. Increase Trimp, O2max indicators, U19, U17 echelon forecasting groups to determine the correct rate went up by 12.5% and 25%; increase HB, CK, T indicators, U19 echelon discrimination accuracy rate increased by 25%, U17 echelon discrimination accuracy rate remains constant; increase HR ＊ VO2max interaction indicators, U19 echelon determines the correct rate of 87.5%, U17 echelon discrimination accuracy rate of 100%. The results show that, Introduction of indicators to monitor the physiological process can be efficient football training feedback body＇s stress response, provide additional support to sports training.