Skip to main content

Table 4 Proportion of well-classified patients with real datasets

From: Importance of data structure in comparing two dimension reduction methods for classification of microarray gene expression data

 

PLS+DA

PCA+DA

BGA

DLBCL

0.51(0.14)//12

0.49(0.09)//13

0.43(0.10)

Prostate

0.97(0.06)//10

0.96(0.07)//9

0.70(0.09)

ALL

0.73(0.05)//10

0.57(0.08)//1

0.60(0.06)

Leukaemia

0.97(0.03)//1

0.95(0.04)//5

0.98(0.03)

  1. Mean (Standard deviation) over the fifty cross-validation runs for the optimal number of component (indicated after //). Results were obtained with the DLBCL, the prostate, the ALL, and the leukaemia datasets. PLS: Partial Least Squares – PCA: Principal Components Analysis – DA: Discriminant Analysis.
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us