基于机器学习的网球最佳回球落点决策模型

Machine Learning-based Decision Model for Optimal Return Placement in Tennis

以网球运动中多拍相持回合的战术分析指导为目标,将网球落点按照区域分布情况进行量化抽象,采集了2022年美网冠军卡洛斯·阿尔卡拉斯7场对局1808盘对局数据,提供了小尺度的自动化网球战术辅助的系统框架,并分别应用支持向量机、神经网络、随机森林等机器学习的训练模型构建多拍相持下的最佳回球落点决策模型。结果表明,随机森林方法的建模效果最佳,选取拍数为5拍时回球落点精度达到76.66%。基于机器学习的网球战术智能量化分析有利于为提高网球运动员技战术水平提供数据与技术支撑。

This study collected 1808 sets data from 7 matches of Carlos Alcaraz who was the champion of 2022 US Open,the tennis ball landing spots were quantitatively abstracted based on regional distribution,and a framework of micro-scale automated tennis tactic assistance was provided.The machine learning models such as Support Vector Machines,Neural Networks,and Random Forests were used to construct an optimal ball landing decision model for multi-stroke rallies.The results showed that the random forest was the most effective method in constructing the decision model,and the peak accuracy can reach to 76.66%for ball landing decisions after 5 strokes.Based on machine learning,the intelligent quantitative analysis can provide data and technical support for the improvement of professional tennis tactics.