摘要
预测体育成绩是制定科学体育训练规划的关键,针对当前模型预测精度低的问题,提出因子分析与神经网络相融合的体育成绩预测模型。根据体育成绩先验信息构建自相似回归模型,对体育成绩数据进行经验模态分解和因子分析,采用BP神经网络建立体育成绩预测模型,并通过仿真实验对性能进行测试,结果表明,采用该模型进行体育成绩预测的精度较高,收敛性较好。
The sports result prediction is the key to formulate the scientific sports training plan. Aiming at the low predic- tion accuracy of the current models, a sports result prediction model fusing the factor analysis with neural network is put for- ward. The self-similarity regression model was constructed according to the sports result apriori information. And then the empiri- cal mode decomposition and factor analysis were conducted for the sports result data. The BP neural network is used to establish the sports result prediction model. The performance of the model was tested with simulation experiments. The results show that the model has higher precision of sports result prediction, and good convergence property.
出处
《现代电子技术》
北大核心
2017年第5期130-133,共4页
Modern Electronics Technique
关键词
经验模态分解
时间序列
因子分析
神经网络
empirical mode decomposition
time series
factor analysis
neural network
