基于优化XGBoost的运动效果量化评估与分析研究

doi:10.15942/j.jcsu.2022.06.014

成都体育学院学报 ›› 2022, Vol. 48 ›› Issue (6): 85-92.doi: 10.15942/j.jcsu.2022.06.014

基于优化XGBoost的运动效果量化评估与分析研究

张晟¹, 刘长江¹, 苗凯尧², 张萌², 刘雨童³

1.西安交通大学体育中心,陕西西安 710049;
2.西安交通大学电信学部,陕西西安 710049;
3.重庆大学自动化学院,重庆 400044

收稿日期:2021-12-22 修回日期:2022-04-29 出版日期:2022-11-15 发布日期:2022-12-01
作者简介:张晟, 硕士,副教授,硕士生导师,研究方向:体育教育与运动训练。E-mail:zhoms@mail.xjtu.edu.cn。
基金资助:
陕西省研究生教育综合改革研究与实践项目 “基于AI的可穿戴运动辅助系统在研究生体育教学中的开发与应用”(YJSZG2020009)。

Research on Quantitative Evaluation and Analysis of Exercise Effect Based on Optimized XGBoost

ZHANG Sheng¹, LIU Changjiang¹, MIAO Kaiyao², ZHANG Meng², LIU Yutong³

1. The Center for Physical Education, Xi'an Jiaotong University,Xi'an Shanxi 710049;
2. The Faculty of Electronic and Information Engineering, Xi'an Jiaotong University,Xi'an Shanxi 710049;
3. The School of Automation, Chongqing University,Chongqing 400044

Received:2021-12-22 Revised:2022-04-29 Online:2022-11-15 Published:2022-12-01

摘要/Abstract

摘要： 科学运动是提升大学生身体综合素质的基础性工程。为有效量化评估大学生速度、耐力等身体素质,本文通过研究不同运动特征对跑步运动效果的影响,基于贝叶斯优化的XGBoost算法提出一种运动效果评估策略,并通过引入SHAP算法来量化特征贡献,分析影响运动效果的重要因素。同时,在运动数据分类挖掘与评估中考虑不同种类运动过程的差异,确保运动效果评估的准确性。最后通过实验验证了基于贝叶斯优化的XGBoost在运动效果评估上的性能,并利用SHAP值来分析影响运动效果的不同运动特征及其重要性。本文提出的算法不仅可用于分析大学生锻炼的运动效果,还能为竞技体育的技能形成与成绩突破、大众健身的体质健康管理,提供更为先进、可靠、精准的数据分析方法。

关键词: 运动效果评估, XGBoost算法, SHAP值, 大学生体育锻炼, 贝叶斯优化

Abstract: Scientific exercise is a fundamental project to improve the comprehensive physical quality of college students. In order to effectively assess the speed and endurance quality of college students, this paper proposes an exercise effect evaluate strategy based on the XGBoost algorithm with Bayesian optimization by studying the effect of different exercise features on the exercise effect, and analyzes the important factors affecting the exercise effect by introducing the SHAP algorithm to quantify the feature contribution. Meanwhile, the differences of different kinds of motion processes are considered in motion data classification mining and evaluation to ensure the accuracy of motion effect evaluation. Finally, the performance of XGBoost based on Bayesian optimization for motion effect assessment is experimentally verified, and SHAP values are used to analyze different motion features and their importance. The algorithm proposed in this paper can not only be used to analyze the exercise effect of college students, but also provide a more advanced, reliable and accurate data analysis method for the formation of skills and breakthroughs in competitive sports, and the physical health management of public fitness.

Key words: exercise effect evaluation, XGBoost algorithm, SHAP value, physical exercise of college student, Bayesian optimization

中图分类号:

G808.12

张晟, 刘长江, 苗凯尧, 张萌, 刘雨童. 基于优化XGBoost的运动效果量化评估与分析研究[J]. 成都体育学院学报, 2022, 48(6): 85-92.

ZHANG Sheng, LIU Changjiang, MIAO Kaiyao, ZHANG Meng, LIU Yutong. Research on Quantitative Evaluation and Analysis of Exercise Effect Based on Optimized XGBoost[J]. Journal of Chengdu Sport University, 2022, 48(6): 85-92.

导出引用管理器 EndNote|Reference Manager|ProCite|BibTeX|RefWorks

链接本文: https://cdtyxb.cdsu.edu.cn/CN/10.15942/j.jcsu.2022.06.014

https://cdtyxb.cdsu.edu.cn/CN/Y2022/V48/I6/85

参考文献

[1] 湖南省人民政府门户网站.习近平总书记在教育文化卫生体育领域专家代表座谈会上的重要讲话在湖南引起热烈反响[EB/OL]. (2020-09-23) [2021-11-18]. http://www.hunan.gov.cn/topic/ljzt/fxrl/202009/t20200929_13772364.html
[2] 胡精超, 王莉. 数据挖掘在大学生体质健康测试中的应用研究[J]. 吉林体育学院学报, 2017,33(3):8-11.
[3] 王廉秀. 健美操和慢跑有氧运动效果的比较研究[J]. 福建体育科技, 2019, 38(2):62-64.
[4] 李佳佳, 丁伟, 王伯伟, 等. 基于随机森林的民俗体育对身体指标影响评估方法[J]. 南京大学学报 (自然科学版), 2021, 57(1):59-67.
[5] 李响. 基于数据挖掘技术的农业院校大学生体能指标的相关性研究[D]. 长春:吉林农业大学, 2019:1-3.
[6] PADHY N, MISHARA P, PANIGRAHI R. The survey of data mining applications and feature scope[J]. International journal of computer science, engineering and information technology,2012,2(3):43-58.
[7] 杨旭, 周明, 周金海, 等. 基于 WPF的运动效果评估系统[J]. 电子设计工程, 2015,23(3):5-7.
[8] 戴敏, 黄亚楼. 基于数据挖掘的运动员体能测试数据分析[J]. 计算机工程与应用, 2003, 39(9):38-40.
[9] 董潇潇, 胡延, 陈彦萍. 关联规则挖掘算法在大学生体育成绩分析中的应用[J]. 自动化与仪器仪表, 2017 (8):225-227.
[10] 陈彩虹. 数据挖掘在学生体质健康测试系统中的应用[J]. 自动化与仪器仪表, 2016 (12):227-228,232.
[11] ZHU S. Research on data mining of education technical ability training for physical education students based on apriori algorithm[J]. Cluster Computing, 2019, 22(6):14811-14818.
[12] MA G, ZHANG L, LI S. ID3-based classification of college students' physical fitness data [M//]. Recent trends in intelligent computing communication and devices. Singapore:Springer, 2020:245-252.
[13] KE G, MENG Q, FINLEY T, et al. Lightgbm:a highly efficient gradient boosting decision tree[C]. Advances in Neural Information Processing Systems, 2017:3146-3154.
[14] CHEN T, GUESTRIN C.XGBoost:A scalable tree boosting system[C]. Proceedings of the 22nd ACM Sigkdd International Conference on Knowledge Discovery and Data Mining, 2016:785-794.
[15] LUNDBERG S M, LEE S I. A unified approach to interpreting model predictions[C]. Proceedings of the 31st International Conference on Neural Information Processing Systems. 2017:4768-4777.
[16] 徐成为. 大学生 “自主锻炼” 现状调查与培养策略[D].烟台:鲁东大学,2017:1-3.
[17] 郭方印, 刘干. 浅谈中长跑的速度耐力训练[J].田径, 2012 (11):7-7.
[18] 沈彤辉. 中国业余马拉松运动员心电图特点与异常发生率研究[D]. 杭州:浙江大学, 2019:22-24.
[19] CASA D, STEARN R. Influence of hydration on physiological function and performance during trail running in the heat[J]. Journal of Athletic training,2010, 45 (2):147-156.
[20] DENIELS J. Daniels' running formular [M]. Champiagn:Human Kinetics, 2013:77-80.
[21] 刘汉锋. 梯度提升决策树性能优化研究[D]. 广州:广东外语外贸大学, 2020:7-8.
[22] FARSEEV A, CHUA T S. Tweetfit:Fusing multiple social media and sensor data for wellness profile learning[C]. Thirty-First AAAI Conference on Artificial Intelligence, 2017:95-101.

基于优化XGBoost的运动效果量化评估与分析研究

Research on Quantitative Evaluation and Analysis of Exercise Effect Based on Optimized XGBoost

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 2

编辑推荐

Metrics

本文评价

[1]	李捷,裘晟,王晓军,李端英,李稚. 论运动训练的科学化与竞技训练实践的几个关键问题(中) ——传统“大周期”训练分期生物逻辑基础初步分析[J]. 成都体育学院学报, 2018, 44(3): 19-24.
[2]	杨群茹,刘建和. 运动训练方法历史演进的几个问题综述[J]. 成都体育学院学报, 2018, 44(3): 108-113.