计算机工程与应用 ›› 2009, Vol. 45 ›› Issue (32): 38-40.DOI: 10.3778/j.issn.1002-8331.2009.32.012

• 研究、探讨 • 上一篇    下一篇

优化KPCA特征提取下的FCM算法研究

蔡静颖,张 永,张凤梅,谢福鼎   

  1. 辽宁师范大学 计算机系,辽宁 大连 116081
  • 收稿日期:2008-11-27 修回日期:2009-02-03 出版日期:2009-11-11 发布日期:2009-11-11
  • 通讯作者: 蔡静颖

Fuzzy C-Mean algorithm based on optimized KPCA feature extraction

CAI Jing-ying,ZHANG Yong,ZHANG Feng-mei,XIE Fu-ding   

  1. Department of Computer,Liaoning Normal University,Dalian,Liaoning 116081,China
  • Received:2008-11-27 Revised:2009-02-03 Online:2009-11-11 Published:2009-11-11
  • Contact: CAI Jing-ying

摘要: 利用核函数主元分析(KPCA)方法对大样本、高维数据进行特征提取预处理,并结合文化算法(CA)选择最优或接近最优的核函数,将其用于模糊C均值(FCM)聚类中,不但有效地提取了样本的非线性信息,而且使样本维数得到约简。实验表明该方法具有较好的聚类效果和更少的训练时间。

关键词: 核函数主元分析, 文化算法, 模糊聚类

Abstract: Kernel PCA method extracts feature from large samples and high dimension data sets,combining CA to select optimized kernel function or near optimized kernel function.FCM based on the method not only effectively extracts the nonlinear information from the samples but also reduces dimension.Experiment shows its better clustering result and less train time.

Key words: kernel principle component analysis, cultural algorithm, fuzzy clustering

中图分类号: