摘要
针对自组织模糊神经网络,提出了一种新的结构辨识算法.通过建立输入和输出相似性准则,提出一个用于提取模糊规则的新型聚类算法.所提方法的显著优点是克服了传统神经网络的维度灾难问题.
New structure identification algorithm for the self-organizing fuzzy neural network is proposed. By consturcting the input and output similarity principle, a new clustering algorithm for extracting fuzzy rules is proposed. One of the main advantages of the proposed approach is that the problem of the curse of dimensionality for the traditional neural networks is overcome.
出处
《河南师范大学学报（自然科学版）》
CAS
CSCD
北大核心
2012年第4期130-132,共3页
Journal of Henan Normal University(Natural Science Edition)
基金
河南省基础与前沿技术研究(112400430087
122300410353)
关键词
神经网络
聚类算法
结构辨识
参数调整
Neural network
clustering algorithm
structure identification
parameter adjustment
