摘要
为了提高运动成绩的预测模型精度,为运动员合理训练提供科学基础,提出基于多因素筛选与权重估计的1 000 m跑成绩预测模型。先构建1 000 m跑成绩统计参数分析模型,获取1 000 m跑成绩预测时的全部影响因素;然后使用基于多因素筛选的1 000 m跑成绩核心预测因素提取方法,在全部影响因素中提取1 000 m跑成绩核心预测因素;最后通过核心预测因素的加权形式完成预测模型的构建。将所构建模型应用于仿真测试中可知,该模型的预测结果和对比模型相比,预测结果和真实成绩吻合度最高,且预测时间最短,可在最少的时间中实现最高精度的1 000 m跑成绩预测。
In order to improve accuracy of prediction model for sport performance and provide scientific basis for athletes′reasonable training,a 1000 m running performance prediction model based on multi⁃factor screening and weight estimation is designed.The statistical parameter analysis model of 1000 m running performance is constructed to obtain all the influencing factors of 1000 m running performance prediction,and then the 1000 m running performance core prediction factor extraction method based on multi⁃factor screening is used to extract core prediction factors of 1000 running results from the influence factors.The prediction model is constructed by weighting the core prediction factors.When the model is applied to the simulation test,it can be seen that,in comparison with the contrast models,the prediction result of the designed model has the highest agreement with the real performance,its prediction time is the shortest,and it can achieve the highest accuracy of 1000 m running performance prediction in the minimum time.
作者
马文新
蔡珊
MA Wenxin;CAI Shan(Xi’an Physical Education University,Xi’an 710068,China;Xi’an University of Science and Technology,Xi’an 710054,China)
出处
《现代电子技术》
2021年第16期133-137,共5页
Modern Electronics Technique
基金
国家自然科学基金(21473132)
陕西省教育厅2019年专项科研计划项目(19JK0510)
教育部高教司2019年第一批产学研协同育人项目(201901044004)。
关键词
1000
m跑
预测模型
多因素筛选
权重估计
预测因素提取
仿真测试
1000 m running
prediction model
multi⁃factor screening
weight estimation
prediction factor extraction
simulation testing
