摘要
提出基于改进K-means聚类的体能训练数据异常识别方法,提高体能训练异常数据识别精度.确定主要的体能训练数据并获取特征参数样本集;利用改进K-means算法进行聚类处理,完成体能训练数据聚类;采用径向基函数构建体能训练数据异常识别模型.实验结果表明,体能训练数据异常识别方法对体能训练异常数据识别的精度明显提升,为体能训练提供可靠的数据支持.
A method for anomaly detection in physical training data based on improved K-means clustering is pro-posed to enhance the accuracy of identifying abnormal data in physical training.The primary physical training data are identified,and a sample set of feature parameters is obtained.The improved K-means algorithm is utilized for clustering to complete the clustering of physical training data.A radial basis function is employed to construct an a-nomaly detection model for physical training data.Experimental results demonstrate that this method for anomaly de-tection in physical training data significantly improves the accuracy of identifying abnormal data,providing reliable data support for physical training.
作者
郑德玲
ZHENG Deling(Shanghai University of Sport,School of Wushu Shanghai,Shanghai 200438,China;Anhui Finance&Trade Vocational College,Public Education Department Hefei,Hefei 230000,China)
出处
《牡丹江师范学院学报(自然科学版)》
2025年第3期63-67,共5页
Journal of Mudanjiang Normal University:Natural Sciences Edition
基金
2022年安徽省高等学校科学研究项目(2022AH052531)
2024年安徽省高校科学研究重点项目(2024AH052143).
关键词
改进K-means聚类算法
体能训练
improved K-means clustering algorithm
physical training
training data
outliers
