Computer Engineering and Applications ›› 2009, Vol. 45 ›› Issue (22): 111-113.DOI: 10.3778/j.issn.1002-8331.2009.22.036

• 数据库、信息处理 • Previous Articles     Next Articles

Imputing microarray missing values with ordering

ZHOU Xiu-mei1,LI Zuo-chun2,QIN Ze2   

  1. 1.Department of Mathematics and Computer Science,Nanning Prefecture Education Institute,Nanning 530001,China
    2.Department of Physics,Nanning Prefecture Education Institute,Nanning 530001,China
  • Received:2009-02-16 Revised:2009-04-02 Online:2009-08-01 Published:2009-08-01
  • Contact: ZHOU Xiu-mei

有序填充微阵列缺失数据

周秀梅1,李作春2,覃 泽2   

  1. 1.南宁地区教育学院 数学与计算机科学系,南宁 530001
    2.南宁地区教育学院 物理系,南宁 530001
  • 通讯作者: 周秀梅

Abstract: The paper firstly analyzes the disadvantages of nearest neighbor based Euclidean distance especially for the microarray missing values with high missing ratio,then proposes an effective imputation algorithm based on Mahalanobis distance metric with ordering.And the experiments in real DNA data show the proposed algorithm outperform the existing algorithms in the terms of RMSE.

Key words: Mahalanobis distance, imputing missing data, imputation ordering

摘要: 针对欧式距离填充算法不足和微阵列数据集中缺失数据比率过大问题,提出了使用马氏距离有序填充微阵列的最近邻算法,能充分使用数据集中所有有效信息填充缺失数据,真实基因数据集的实验结果显示改进后的最近邻算法明显优于存在算法。

关键词: 马氏距离, 缺失数据填充, 填充顺序