中国组织工程研究 ›› 2012, Vol. 16 ›› Issue (37): 6956-6960.doi: 10.3969/j.issn.2095-4344.2012.37.024

• 组织构建基础实验 basic experiments in tissue construction • 上一篇    下一篇

构建人工神经网络体质综合评价模型

张崇林1, 2,虞丽娟1,吴卫兵1   

  1. 1上海体育学院运动科学学院,上海市 200438
    2井冈山大学体育学院,江西省吉安市 343009
  • 收稿日期:2012-06-04 修回日期:2012-07-11 出版日期:2012-09-09 发布日期:2012-09-09
  • 通讯作者: 虞丽娟,博士,教授、博士生导师。上海体育学院运动科学学院,上海市200438 ljyu@shou.edu.cn
  • 作者简介:张崇林☆,1976年生,男,湖北省孝感市人,博士,2012年上海体育学院毕业,讲师。 ccnuzcl@163.com

Modeling of a comprehensive fitness evaluation model based on artificial neural network

Zhang Chong-lin1, 2, Yu Li-juan1, Wu Wei-bing1   

  1. 1School of Kinesiology of Shanghai University of Sport, Shanghai 200438, China
    2Institute of Sports of Jinggangshan University, Ji’an 343009, Jiangxi Province, China
  • Received:2012-06-04 Revised:2012-07-11 Online:2012-09-09 Published:2012-09-09
  • Contact: Yu Li-juan, Doctor, Professor, Doctoral supervisor, School of Kinesiology of Shanghai University of Sport, Shanghai 200438, China ljyu@shou.edu.cn
  • About author:Zhang Chong-lin☆, Doctor, Lecturer, School of Kinesiology of Shanghai University of Sport, Shanghai 200438, China; Institute of Sports of Jinggangshan University, Ji’an 343009, Jiangxi Province, China ccnuzcl@163.com

摘要:

背景:诸多学者都认同不同的体质指标对体质综合评价作用不同,却一直没有有效方法建立体质综合评价模型。
目的:采用人工神经网络技术建立体质综合评价模型。
方法:以上海市某高校教职工为研究对象,以人工神经网络技术,建立的不同年龄、不同性别的二级指标、三级指标体质综合评价模型。
结果与结论:实验所建立的二级指标、三级指标体质综合评价模型的拟合度分别>93%和>94%,权重计算结果可信。在此模型中,反映身体素质的指标所占的权重系数总体最高,其次为反映身体形态的指标,而反映身体功能的指标较低。说明实验采用人工神经网络技术成功建立体质综合评价模型。

关键词: 人工神经网络, 体质, 综合评价, 模型, 组织构建

Abstract:

BACKGROUND: Many scholars agree that different fitness indexes play different roles in the comprehensive evaluation of the fitness. But there is no effective way to build a comprehensive fitness evaluation model.
OBJECTIVE: To establish the comprehensive fitness evaluation model based on artificial neural network.
METHODS: The university faculties in Shanghai were selected as objects and the artificial neural network technology was used to establish the comprehensive fitness evaluation models of second-rank index and third-rank index in different ages ad genders.
RESULTS AND CONCLUSION: The fitting degrees of the second-rank index and third-rank index comprehensive fitness evaluation models established in this experiment were > 93% and > 94%. The weight calculation results are reliable. In this model, the indicators reflecting body diathesis accounted for the highest proportion of weight coefficients, followed by the indicators reflecting the body shape, and the indicators reflecting the physical function accounted for the lowest proportion of weight coefficients. The findings confirmed that the comprehensive fitness evaluation model was successfully established by using the artificial neural network technology.

中图分类号: