文献类型：期刊文献

中文题名：基于深度学习的大学生体质监测成绩预测研究

英文题名：Research on the Prediction of College Students’ Physique Monitoring Performance based on Deep Learning

作者：黄彩云[1];何吉福[1];胡艺[1];王楠[1];陈沛[1]

第一作者：黄彩云

机构：[1]甘肃中医药大学,甘肃兰州730000

第一机构：甘肃中医药大学

年份：2022

卷号：30

期号：6

起止页码：146

中文期刊名：体育科技文献通报

外文期刊名：Bulletin of Sport Science & Technology

收录：国家哲学社会科学学术期刊数据库

基金：甘肃中医药大学科技创新项目《功能性动作筛查在大学生体质健康测试中的应用研究》(项目编号:30740301)。

语种：中文

中文关键词：体质监测;成绩预测;人工智能;深度学习;神经网络模型

外文关键词：physique monitoring;performance prediction;artificial intelligence;deep learning;neural network model

摘要：针对目前缺乏有效预测大学生体质健康状况手段的问题,本文提出一种基于深度学习的大学生体质监测成绩预测方法。该方法借助目前热门的人工智能深度学习技术,设计了一种深度全连接神经网络模型,以甘肃某高校近三年来新入学学生体质监测数据作为训练和测试数据集,对深度全连接神经网络模型进行了训练和测试。结果表明,该神经网络模型对预测大学生50米跑和立定跳远等体质监测成绩有较高的精准度,准确率高于传统的多元线性回归方法,并且可有效的避免在操场实施体质监测过程中的一些弊端,对提前了解学生的体质健康状况及合理安排体育锻炼具有一定的实践参考价值。
In view of the lack of effective methods to predict the physique health status of college students, this paper proposed a physique monitoring performance prediction method based on deep learning.Using currently popular deep learning technology of artificial intelligence, this research designed a deep fully-connected neural network model.Using the physical monitoring data of newly admitted students in a college in Gansu in the past three years as the training and test data sets, the deep fully-connected neural network model is carried out and tested.The results show that the neural network model has high accuracy in predicting college students’ 50-meter running and standing long jump performances, and the accuracy is higher than the traditional multiple linear regression method, and can effectively avoid some disadvantages in the physical fitness monitoring process in the playground.This method has certain practical reference value for understanding the physical health status of students in advance and arranging physical exercise reasonably.