论文部分内容阅读
给出了一种新型的模糊神经网络模型 .该模型不需要领域专家的知识进行指导 ,而是通过对样本竞争分类产生模糊规则 .每类样本对应于一条模糊规则 ,每条模糊规则的后件部分为一个对本类样本进行过学习训练的神经网络 .文章以模糊神经网络在时间序列分析中的应用为例 ,通过与传统的时间序列分析方法以及前向神经网络方法的对比 ,说明了新型模糊神经网络的有效性 .
A new model of fuzzy neural network is proposed, which does not need the knowledge of experts in the field to guide but produces fuzzy rules by competing the samples. Each type of sample corresponds to a fuzzy rule, and the back part of each fuzzy rule Part of which is a neural network which has been trained in this kind of samples.This paper takes the application of fuzzy neural network in time series analysis as an example to illustrate the new fuzzy method by comparing with the traditional time series analysis method and the forward neural network method The effectiveness of neural networks.