| 确定前向神经网络隐层节点数的模糊聚类分析法 |
| |
| 引用本文： | 许世刚,索丽生.确定前向神经网络隐层节点数的模糊聚类分析法[J].河海大学学报(自然科学版),2001,29(3):17-20. |
| |
| 作者姓名： | 许世刚 索丽生 |
| |
| 作者单位： | 河海大学水利水电工程学院, |
| |
| 摘 要： | 概述了前向网络结构目前的研究状况，从隐层的本质特征出发，提出对样本数据采用模糊聚类分析法进行特征抽取，并综合考虑了样本的输入、输出信息，实例分析说明此方法可行的，较互相关分析法、奇异值分解法有更高的效率。
|
| 关 键 词： | 前向神经网络 模糊聚类 奇异值分解 |
| 文章编号： | 1000-1980(2001)03-0017-04 |
| 修稿时间： | 2000年5月9日 |
Fuzzy Cluster Analysis Method for Determining the Number of Hidden Nodes of Feedforward Neural Networks |
| |
| XU Shi-gang,SUO Li-sheng.Fuzzy Cluster Analysis Method for Determining the Number of Hidden Nodes of Feedforward Neural Networks[J].Journal of Hohai University (Natural Sciences ),2001,29(3):17-20. |
| |
| Authors: | XU Shi-gang SUO Li-sheng |
| |
| Abstract: | In this paper, advances in architecture of feedforward networks are surveyed. Based on the essence of the hidden layer,a fuzzy cluster analysis method is put forward to extract the feature of sample data, moreover, the input and output information is comprehensively considered. An example has proved the feasibility of the fuzzy cluster analysis methed,which is more efficient than the methods of interrelation analysis and singular value decomposition. |
| |
| Keywords: | feedforward neural networks hidden nodes extraction of feature fuzzy cluster interrelation singular value decomposition |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
| 点击此处可从《河海大学学报(自然科学版)》浏览原始摘要信息 |
| 点击此处可从《河海大学学报(自然科学版)》下载免费的PDF全文 |
