南京大学学报(自然科学版) ›› 2021, Vol. 57 ›› Issue (1): 59–67.doi: 10.13232/j.cnki.jnju.2021.01.007

• • 上一篇    下一篇

基于随机森林的民俗体育对身体指标影响评估方法

李佳佳1, 丁伟2(), 王伯伟1, 聂秀山3, 崔超然1   

  1. 1.山东财经大学计算机科学与技术学院,济南,250014
    2.山东财经大学体育学院,济南,250014
    3.山东建筑大学计算机科学与技术学院,济南,250101
  • 收稿日期:2020-10-09 出版日期:2021-01-21 发布日期:2021-01-21
  • 通讯作者: 丁伟 E-mail:scdw2013@163.com
  • 作者简介:E⁃mail: scdw2013@163.com
  • 基金资助:
    国家社会科学基金(17BTY056);山东省科技发展计划(2011GGH21816)

An evaluation method for the influence of folk sports on body indicators based on random forest

Jiajia Li1, Wei Ding2(), Bowei Wang1, Xiushan Nie3, Chaoran Cui1   

  1. 1.School of Computer Science and Technology,Shandong University of Finance and Economics,Ji'nan,250014,China
    2.Physical Education College,Shandong University of Finance and Economics, Ji'nan,250014,China
    3.School of Computer Science and Technology,Shandong Jianzhu University,Ji'nan,250101,China
  • Received:2020-10-09 Online:2021-01-21 Published:2021-01-21
  • Contact: Wei Ding E-mail:scdw2013@163.com

摘要:

随着体育事业的发展,传统体育项目越来越受重视.作为我国传统体育运动的重要组成部分,民俗体育运动的发展显得尤为重要.为研究民俗体育项目对健身效果的影响,组织多个对象分别进行一段时间内的角力类、竞足类和技巧类三类民俗体育运动训练,并观察对象训练后在身体形态、身体机能、身体素质三方面共计32个代表性身体指标的变化情况.进一步,以身体指标变化情况作为特征表示观察对象,采用随机森林算法预测不同对象在训练阶段进行民俗体育运动的种类,在过程中基于信息增益进行特征选择,从而度量不同类别民俗体育运动对各项身体指标的影响程度.将获得的不同运动对各项身体指标的影响程度与真实影响情况进行评估分析,揭示民俗体育运动与人身体机能的关联关系.此外,实验结果证明,和基准线算法相比,提出的算法有更高的预测准确性.

关键词: 民俗体育, 随机森林, 身体指标, 数据挖掘, 特征选择

Abstract:

With the development of sports,more and more attention has been paid to traditional sports. As an important part of Chinese traditional sports,the development of folk sports is particularly important. In order to study the influence of folk sports on the effect of physical fitness,this article organizes a number of subjects to conduct three types of folk sports training including wrestling,foot races and skills sports for a period of time,and observes a total of 32 representative body indicators in three aspects: body shape,physical function and physical fitness after training. As for a further comment,based on the changes of indicators,this article uses changes in physical indicators as features to represent individuals. Random forest algorithm is used to predict what kind of folk sports different objects will perform in the training phase. In the process,feature selection is based on information gain to measure impact. Thus,it measures the influence degree of different types of folk sports on body indicators. In this paper,the influence degree of different folk sports on body indicators and actual impact are evaluated and studied to reveal the correlation between folk sports and human body functions. In addition,the experimental results show that the proposed algorithm has higher prediction accuracy than the baseline algorithm.

Key words: folk sports, random forest, body indicators, data mining, feature selection

中图分类号: 

  • TP391

图1

基于随机森林的民俗体育对身体指标影响评估算法模型示意图"

表1

Folk?CS数据库的测量数据举例"

Person身高(cm)坐高(cm)转肩(cm)50 m短跑(s)平均握力(kg)背肌力(kg)纵横差(cm)
P01151.581.368.49.520.6560.219.4
P02154.983.263.592548.918.8
P03148.679.477.59.219.7551.228.85
P04155.384.364.58.922.658.520.9
P11153.281.970.49.225.1565.718.8
P12156.48465.28.828.655.218.2
P13150.380.278.78.924.1557.128.25
P14156.885.166.28.726.764.420.3

表2

各类运动对身体指标影响程度的真实排序"

12345678910
角力类背肌力平均握力

上臂围

松紧差

立位转体心功指数呼吸差

一分钟

仰卧起坐

肺活量俯卧撑体脂率
竞足类往返跑反复横跨立位转体

十字

变向跑

50 m

短跑

立定跳远心功指数

闭目

单足立

俯卧撑纵横叉
技巧类

选择

反应时

反复横跨心功指数

闭目

单足立

50 m

短跑

1 min

抛网球

十字

变向跑

立位转体往返跑纵横叉

表3

Folk?CS数据库的示例"

Person身高坐高转肩50 m短跑平均握力背肌力纵横差
C110.0112211220.007380.029240.0315789470.2187294270.0913621260.03145
C120.0096836670.0096150.0267720.0222222220.1440207390.1288343560.032537
C130.0114401080.0100760.0154840.0326086960.223416290.1152343750.021081
C140.0096587250.009490.0263570.022471910.1815756430.1008547010.029159

图2

基于随机森林的民俗体育对身体指标影响评估算法流程示意图"

图3

角力类运动影响最大的20个身体指标"

图4

竞足类运动影响最大的20个身体指标"

图5

技巧类运动影响最大的20个身体指标"

表4

五种方法在Folk?CS数据集上的分类性能比较"

方法角力类竞足类技巧类
top@3top@5top@10top@3top@5top@10top@3top@5top@10
均值0.330.40.700.20.40.330.40.3
方差0.330.20.7000.400.20.3
SVM00.40.60.330.60.7000.7
RFE00.40.500.20.400.20.5
Our?RF0.330.60.50.330.60.70.330.80.7
1 林顺治. 体育强国建设中民俗体育文化发展研究. 军事体育进修学院学报, 2010,29(1):26-29.
Lin S Z. Researching on the development of folk?custom sports culture under a world sports power construction. Journal of PLA Institute of Physical Education,2010,29(1):26-29.
2 Powers S K,Dodd S L. Total fitness:exercise,nutrition,and wellness. Boston:Allyn and Bacon,1999,398.
3 许浩,姜文凯. 对江苏省成年人BMI指数的研究. 体育与科学,2001,22(6):39-43.
Xu H,Jiang K W. A study of the body mass index of adults in Jiangsu Province. Journal of Sports and Science,2001,22(6):39-43.
4 张勇,蒋月玲,钱永东. 2015年浙江师范大学男性教职工体质状况研究. 当代体育科技,2015,5(35):129-130.
Zhang Yo,Jiang Y L,Qian Y D. Study on physical fitness of male faculty in Zhejiang Normal University in 2015. Contemporary Sports Technology,2015,5(35):129-130.
5 周皎,乔京转,马慧敏. 生活方式对大学生体质健康的影响研究. 辽宁体育科技,2015,37(6):67-69.
Zhou J,Qiao J Z,Ma H M. A Study on the life style influencing undergraduates’ physical health. Liaoning Sport Science and Technology,2015,37(6):67-69.
6 乔克满,孙卫. 身体形态、机能、素质指标的关联分析. 周口师范学院学报,2010,27(5):144-146.
Qiao K M,Sun W. Association analysis on indices of body pattern,function and content. Journal of Zhoukou Normal University,2010,27(5):144-146.
7 陶弥锋. 江西高校学生体质健康现状及主要问题分析. 当代体育科技,2015,5(35):101-102.
Tao M F. Analysis of the current situation and main problems of college students' physical health in Jiangxi Province. Contemporary Sports Technology,2015,5(35):101-102.
8 杜志峰,周艳华. 中小学校学生体质健康大数据技术应用研究. 青少年体育,2020(4):37-38.
Du Z F,Zhou Y H. Application research on big data technology of physical health of students in primary and middle schools. Youth Sports,2020(4):37-38.
9 张京舒,闫晓晋,胡佩瑾等. 1985—2014年中国汉族13~18岁学生体质健康达标优良率变化趋势及相关因素分析. 中华预防医学杂志,2020,54(9):981-987.
Zhang J S,Yan X J,Hu P J,et al. Trends of prevalence of excellent health status and physical fitness among Chinese Han students aged 13 to 18 years from 1985 to 2014. Chinese Journal of Preventive Medicine,2020,54(9):981-987.
10 Aoki R Y S,Assuncao R M,De Melo P O S,et al. Luck is hard to beat:the difficulty of sports prediction∥Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York,NY,USA:ACM,2017:1367-1376.
11 Decroos T,Bransen L,Van Haaren J,et al. Actions speak louder than goals:Valuing player actions in soccer∥Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. New York,NY,USA:ACM,2018:1851-1861.
12 宋兆铭,叶菁,董如军. 数据挖掘:C5.0决策树算法在警察院校学生体质分析中的应用. 四川体育科学,2020,39(1):52-55,74.
Song Z K,Ye J,Dong R J. Ta mining:application of C5.0 decision tree algorithm in physique analysis of police college. Sichuan Sports Science,2020,39(1):52-55,74.
13 冯敏,冯辉,张一雨等. 随机森林模型分析大学生体质健康影响因素:来源于同济大学568名学生的问卷调查. 中国组织工程研究,2019,23(23):3722-3728.
Feng M,Feng H,Zhang Y Y,et al. Factors influencing physical health of college students analyzed by random forest model:questionnaires of 568 students from Tongji University. Chinese Journal of Tissue Engineering Research,2019,23(23):3722-3728.
14 刘玉琪. 基于随机森林算法的人体运动模式识别研究. 硕士学位论文. 北京:北京邮电大学,2018.
Liu Y Q. Human motion pattern recognition research based on Random Forest algorithm. Master Dissertation. Beijing:Beijing University of Posts and Telecommunications,2018.
15 周志华. 机器学习. 清华大学出版社,2016,74-76.
16 Zhou Z H,Feng J. Deep forest:towards an alternative to deep neural networks. 2017,arXiv:1702.08835.
17 葛绍林,叶剑,何明祥. 基于深度森林的用户购买行为预测模型. 计算机科学,2019,46(9):190-194.
Ge S L,Ye J,He M X. Prediction model of user purchase behavior based on deep forest. Computer Science,2019,46(9):190-194.
18 蒋鹏飞,魏松杰. 基于深度森林与CWGAN?GP的移动应用网络行为分类与评估. 计算机科学,2020,47(1):287-292.
Jiang P F,Wei S J. Classification and evaluation of mobile application network behavior based on Deep Forest and CWGAN?GP. Computer Science,2020,47(1):287-292.
19 朱晓妤,严云洋,刘以安等. 基于深度森林模型的火焰检测. 计算机工程,2018,44(7):264-270.
Zhu X Y,Yan Y Y,Liu Y A,et al. Flame detection based on deep forest model. Computer Engineering,2018,44(7):264-270.
20 韩慧,王黎明,柴玉梅等. 基于强化表征学习深度森林的文本情感分类. 计算机科学,2019,46(7):172-179.
Han H,Wang L M,Chai Y M,et al. Text sentiment classification based on deep forests with enhanced features. Computer Science,2019,46(7):172-179.
21 于剑. 机器学习:从公理到算法. 北京:清华大学出版社,2017,48-51,156-162,170-173.
22 吕世聘,王秀坤,孙岩等. 改进的支持向量机特征选择算法. 计算机工程,2009,35(1):171-172.
Lü S P,Wang X K,Sun Y,et al. Improved feature selection algorithm for SVM. Computer Engineering,2009,35(1):171-172.
[1] 刘琼, 代建华, 陈姣龙. 区间值数据的代价敏感特征选择[J]. 南京大学学报(自然科学版), 2021, 57(1): 121-129.
[2] 陈石,张兴敢. 基于小波包能量熵和随机森林的级联H桥多电平逆变器故障诊断[J]. 南京大学学报(自然科学版), 2020, 56(2): 284-289.
[3] 程玉胜,陈飞,庞淑芳. 标记倾向性的粗糙互信息k特征核选择[J]. 南京大学学报(自然科学版), 2020, 56(1): 19-29.
[4] 刘亮,何庆. 基于改进蝗虫优化算法的特征选择方法[J]. 南京大学学报(自然科学版), 2020, 56(1): 41-50.
[5] 吴静怡,吴钟强,商琳. 基于Shapelet的不相关情感子序列挖掘方法[J]. 南京大学学报(自然科学版), 2020, 56(1): 57-66.
[6] 刘 素, 刘惊雷. 基于特征选择的CP-nets结构学习[J]. 南京大学学报(自然科学版), 2019, 55(1): 14-28.
[7] 胡 淼,王开军,李海超,陈黎飞. 模糊树节点的随机森林与异常点检测[J]. 南京大学学报(自然科学版), 2018, 54(6): 1141-1151.
[8] 陈海娟,冯 翔,虞慧群. 基于预测算子的GSO特征选择算法[J]. 南京大学学报(自然科学版), 2018, 54(6): 1206-1215.
[9] 温 欣1,李德玉1,2*,王素格1,2. 一种基于邻域关系和模糊决策的特征选择方法[J]. 南京大学学报(自然科学版), 2018, 54(4): 733-.
[10] 靳义林1,2*,胡 峰1,2. 基于三支决策的中文文本分类算法研究[J]. 南京大学学报(自然科学版), 2018, 54(4): 794-.
[11]  芦俊丽1,2,王丽珍1*,赵家松1,肖 清1.  从动态空间数据库中挖掘共生关系和竞争关系[J]. 南京大学学报(自然科学版), 2018, 54(2): 436-.
[12]  杨 宇1,吉根林1*,赵 斌1,黄潇婷2.  一种新的基于时空轨迹的汇合模式挖掘算法[J]. 南京大学学报(自然科学版), 2018, 54(1): 97-.
[13]  董利梅,赵 红*,杨文元.  基于稀疏聚类的无监督特征选择[J]. 南京大学学报(自然科学版), 2018, 54(1): 107-.
[14]  崔 晨,邓赵红*,王士同.  面向单调分类的简洁单调TSK模糊系统[J]. 南京大学学报(自然科学版), 2018, 54(1): 124-.
[15]  李 婵,杨文元*,赵 红.  联合依赖最大化与稀疏表示的无监督特征选择方法[J]. 南京大学学报(自然科学版), 2017, 53(4): 775-.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!