计算机工程与应用 ›› 2009, Vol. 45 ›› Issue (25): 52-54.DOI: 10.3778/j.issn.1002-8331.2009.25.016

• 研究、探讨 • 上一篇    下一篇

改进的OLS算法选择RBFNN中心的方法

郑明文   

  1. 中国石油大学(华东) 计算机与通信工程学院,山东 东营 257061
  • 收稿日期:2008-05-13 修回日期:2008-08-25 出版日期:2009-09-01 发布日期:2009-09-01
  • 通讯作者: 郑明文

RBFNN center choice method based on Kohonen network and OLS algorithm

ZHENG Ming-wen   

  1. Department of Computer and Communication Engineering,China University of Petroleum,Dongying,Shandong 257061,China
  • Received:2008-05-13 Revised:2008-08-25 Online:2009-09-01 Published:2009-09-01
  • Contact: ZHENG Ming-wen

摘要: 提出了一种优化选择径向基神经网络数据中心的算法,该算法结合了Kohonen网络的模式分类能力,将初步分类结果用做RBFNN的初始数据中心,然后采用OLS算法进行优化选择,对比仿真实验表明该算法效果比单独使用OLS算法生成的RBFNN性能更好。

关键词: RBF神经网络(RBFNN), 数据中心, Kohonen 网络, 正交最小二乘法

Abstract: This article proposes an optimized choice radial basis function neural network data central algorithm.This algorithm unifies the Kohonen network’s pattern classification ability,classifies firstly the result to make RBFNN the initial data center,and then uses the OLS algorithm to carry on optimized choice.The contrast simulation experiments indicate that this algorithm produces better RBFNN performance than using OLS algorithm independently.

Key words: Radial Basis Function Neural Network(RBFNN), data center, Kohonen network, Orthogonal Least Squares(OLS) method

中图分类号: