建立了神经网络物元模型，并详细分析神经网络物元的可拓性以及基本的物元变换；
Matter-element model of neural networks is constructed. Its extension character and matter-element transformation are detailedly analyzed.
聚类分析神经网络的输入样本序列具有各种不同的空间分布性态，这就要求采用不同的相似匹配准则。
The input series of cluster neural networks have various characteristics in the samples' space, which need corresponding similarity matching rules used in the networks.
通过对背向滑步推铅球技术的实例研究，证明本研究建立的运动技术分析神经网络模型的精度是可靠的，分析方法是可行的。
ANNAMT was constructed in the practical study of the glide-style shot-put. The research testified that the forecasting ability of this model is high and the method of constructing ANNAMT is feasible.
应用推荐