Scholarlink

This study investigates the use of machine learning to predict long jump performance based on biomechanical features. Data from nineteen biomechanical variables collected from elite long jumpers were analyzed, with machine learning models including Random Forest, CatBoost, Gradient Boosting, and XGBoost used for performance prediction. Feature importance analysis, conducted using the Random Forest model, highlighted key predictors such as gender, horizontal velocities during critical strides leading up to take-off, and vertical velocity at take-off. Among the models tested, XGBoost demonstrated the highest accuracy, achieving an \(\text {R}^{2}\) of 0.9454, MAE of 0.1552, and RMSE of 0.1791. These findings emphasize the pivotal role of biomechanical factors in predicting performance and offer data-driven insights for enhancing long jump outcomes for athletes and coaches.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Machine Learning Prediction of Long Jump Performance Based on Biomechanical Factors

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Marouane Baadi,
  • Lekbir Afraites,
  • Soufiane Lyaqini

摘要

This study investigates the use of machine learning to predict long jump performance based on biomechanical features. Data from nineteen biomechanical variables collected from elite long jumpers were analyzed, with machine learning models including Random Forest, CatBoost, Gradient Boosting, and XGBoost used for performance prediction. Feature importance analysis, conducted using the Random Forest model, highlighted key predictors such as gender, horizontal velocities during critical strides leading up to take-off, and vertical velocity at take-off. Among the models tested, XGBoost demonstrated the highest accuracy, achieving an \(\text {R}^{2}\) of 0.9454, MAE of 0.1552, and RMSE of 0.1791. These findings emphasize the pivotal role of biomechanical factors in predicting performance and offer data-driven insights for enhancing long jump outcomes for athletes and coaches.