基于历史数据驱动的运动员成绩估计研究

Research on Performance Estimation of Athletes Based on Historical Data

运动员成绩存在大量的历史数据有待挖掘,为此提出了基于历史数据驱动的运动员成绩估计方法,提升运动员成绩估计质量。利用K最近邻分类算法预处理运动员历史成绩数据集,采用支持向量回归对处理运动员历史成绩数据进行训练,并采用粒子群优化算法优化支持向量机的相关参参数,提升运动员成绩估计的准确性。实验证明所研究方法可有效估计运动员成绩,适用于不同运动项目运动员,成绩估计精度高、估计效率快;对于不同的运动员数量,所研究方法的误差评价指标最低,估计精度最高,估计质量高。

There are a lot of historical data of athletes’performance to be mined,hence this paper puts forward a method of athletes’performance estimation based on historical data to improve the quality of athletes’performance estimation.The K-nearest neighbor classification algorithm is used to preprocess the athletes’historical performance data set.Support vector regression(SVR)is used to train athletes’historical performance data,and particle swarm optimization(PSO)is used to optimize the parameters of support vector machine(SVM)to improve the accuracy of athletes’performance estimation.The results show that the proposed method can effectively estimate the performance of athletes with different sports the performance estimation accuracy is always high,the estimation efficiency is fast.Comparing with different methods,the error evaluation index of the research method is the lowest,the estimation accuracy is the highest,hence,the estimation quality is high.