计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (4): 166-168.

• 数据库与信息处理 • 上一篇    下一篇

基于ICA与SOM的不完整数据处理

奉国和1,5,彭红毅2,蒋春福3,杜 明4   

  1. 1.华南理工大学 计算机科学与工程学院,广州 510640
    2.华南农业大学 理学院 统计系,广州 510642
    3.深圳大学 数学与计算科学学院,广东 深圳 518060
    4.南雄市地税局,广东 韶关 512400
    5.华南师范大学 经济管理学院 信息管理系,广州 510006
  • 收稿日期:2007-06-04 修回日期:2007-08-05 出版日期:2008-02-01 发布日期:2008-02-01
  • 通讯作者: 奉国和

Handling of incomplete data sets based on ICA and SOM

FENG Guo-he1,5,PENG Hong-yi2,JIANG Chun-fu3,DU Ming4   

  1. 1.College of Computer Science & Engineering,South China University of Technology,Guangzhou 510640,China
    2.College of Science,South China Agricultural University,Guangzhou 510642,China
    3.Department of Mathematics,Shenzhen University,Shenzhen,Guangdong 518060,China
    4.Local Tax Bureau of Nanxiong City,Shaoguan,Guangdong 512400,China
    5.College of Economics and Management,South China Normal University,Guangzhou 510006,China
  • Received:2007-06-04 Revised:2007-08-05 Online:2008-02-01 Published:2008-02-01
  • Contact: FENG Guo-he

摘要: 介绍了数据挖掘中不完整数据的研究现状及ICA与SOM的特点,提出了基于ICA与SOM的不完整数据的处理模型IVS-IDH,研究了数据之间存在相关关系且为非高斯分布时不完整数据的处理方法,在SOM基础上取得了不完整数据集的可视化分析结果,从而克服了Wang S提出的不完整数据处理方法的不足。

关键词: 不完整数据, ICA, SOM, 相关关系, 高斯分布

Abstract: The paper introduces the state of incomplete data as well as ICA’s and ViSOM’s characteristics,studies the method of incomplete data sets under the circumstances of that data remain dependent and non-Gaussian.Based on ICA and SOM,a model named IVS-IDH,is proposed in this paper.The proposed model can achieve the visualization of incomplete data sets based on SOM,so that it overcomes the remedy for handing of incomplete data proposed by Wang S.

Key words: incomplete data, Independent Component Analysis(ICA), Self-Organizing Maps(SOM), correlation, Gaussian distribution