【24h】

A New Approach Of Fuzzy Neural Networks

机译:模糊神经网络的一种新方法

获取原文
获取原文并翻译 | 示例

摘要

In this paper, we develop a new approach to obtain fuzzy rule bases and present an improved fuzzy neural networks (FNN) model which consists of four layer neurons. The fuzzy rule bases could be extracted adaptively and directly from training sampled data, and the weights of fuzzy rules reflects the supported grades of the linguistic variables' value of each input to the antecedent fuzzy rules and the important degree of the fuzzy rule in fuzzy rule bases. In the proposed model, the structure and all the parameters such as the membership functions, the weights and so the number of fuzzy bases etc. would be adjusted optimally through back-propagation learning algorithm.
机译:在本文中,我们开发了一种获取模糊规则库的新方法,并提出了一种由四层神经元组成的改进的模糊神经网络(FNN)模型。可以从训练样本数据中自适应地直接提取模糊规则库,模糊规则的权重反映了每个输入对先前模糊规则所支持的语言变量值的支持等级以及模糊规则在模糊规则中的重要程度基地。在该模型中,通过反向传播学习算法可以对结构和所有参数(如隶属函数,权重,模糊基数等)进行最佳调整。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号