摘要
小波神经网络是将小波理论和神经网络理论结合起来的一种神经网络,它避免了BP神经网络结构设计的盲目性和局部最优等非线性优化问题,大大简化了训练,具有较强的函数学习能力和推广能力及广阔的应用前景.根据这一预测模型,运用混合编程技术所开发的分布式运动成绩预测系统在应用中取得了良好的效果.
The wavelet neural network is a kind of neural networks,which can closely combine wavelet theory with neural network theory,and avoids the blindness of framework designs for BP neural networks and the problem of nonlinear optimizations,such as local optimization.So it can greatly simplify the training of neural networks.It has good abilities of function learning and dissemination with a vast range of prospects for application.Applying the distributed training grade prediction system is realized by mix-coding technology,according to this prediction model,obtained satisfied results.
出处
《湖南城市学院学报(自然科学版)》
CAS
2011年第4期69-72,共4页
Journal of Hunan City University:Natural Science
关键词
小波神经网络
运动成绩预测
专家系统
实现技术
wavelet neural network
training grade prediction
expert system
realization technology