微型电脑应用2025，Vol.41Issue(11)：39-42,4.
基于改进SMOTE算法的体育运动录入系统的不平衡数据分类研究分析
Research and Analysis of Unbalanced Data Classification in Sport Input System Based on Improved SMOTE Algorithm
摘要
Abstract
Unbalanced sports data performance has dual characteristics of high dimensionality and unbalanced distribution.For the unbalanced classification problem,most traditional classification models often require a large number of labeled samples,re-sulting in low classification efficiency for unbalanced data.Therefore,it is proposed to use the SMOTE algorithm and the nor-mal distribution idea to replace the uniform random distribution in the original SMOTE algorithm with a normal random distri-bution,so that the newly generated sample points are distributed with a higher probability near the center of a few samples,avoiding the marginalization of extended data and improving the efficiency of unbalanced data classification.The experimental results show that as the oversampling multiple increases,the accuracy,recall,and F1 score show a trend of first increasing and then decreasing.When the oversampling multiple is 1 and 5,the changes in evaluation indicators such as accuracy are relatively small.When the oversampling multiple is 10,the increases in accuracy,recall,and F1 score are significantly obvious,with a maximum accuracy of 97.77％.When the unbalance ratio is from 1∶20 becomes 1∶30,the accuracy of the improved SMOTE al-gorithm decreases from 99.08％to 98.67％.When the unbalance ratio is very large,the improved SMOTE algorithm still has high classification accuracy.关键词
SMOTE算法/正态分布/采样倍数/分类精度Key words
SMOTE algorithm/normal distribution/sampling multiple/classification accuracy分类
信息技术与安全科学引用本文复制引用
姚继富,黄鹏霖..基于改进SMOTE算法的体育运动录入系统的不平衡数据分类研究分析[J].微型电脑应用,2025,41(11):39-42,4.基金项目
山东省教育厅教学研究课题(2023JXY166) （2023JXY166）
