Skip to main content

Table 3 Descriptions of real-life datasets. N is the number of features after preprocessing. K is the number of classes in the dataset.

From: Characteristics of predictor sets found using differential prioritization

Dataset Type N K Training set size:Test set size
BRN cDNA 7452 14 174:83
GCM Affymetrix 10820 14 144:54
NCI60 cDNA 7386 8 40:20
PDL Affymetrix 12011 6 166:82
Lung Affymetrix 1741 5 135:68
SRBC cDNA 2308 4 55:28
MLL Affymetrix 8681 3 48:24
AML/ALL Affymetrix 3571 3 48:24