Skip to main content

Table 5 The best C for ECOC and classifiers based on linear SVMs

From: Multiclass classification of microarray data samples with a reduced number of genes

Dataset

ECOC at η= 5a

ECOC at η= 10a

ECOC at η= 15a

OAAa

Lymphoma

NA

1:1-1

1:1-1

1:1-1

SRCBT

1:1-1

1:1-1

1:1-1

1:1-1

Brain

1:1-1

1:1-1

1:1-1

1:1-1

NCI60

1:1-1

1:1-1

1:1-1

1:0.5-1

Staunton

1:1-1

1:1-1

1:1-1

1:0.5-1

GCMRM

1:1-1

1:1-1

1:1-1

1:1-1

Su

1:1-1

1:1-1

1:1-1

1:0.5-1

GCM

1:1-1

1:1-1

1:1-1

1:0.5-1

  1. The best C for ECOC classifiers of size at most ⌈η·log2M⌉ and OAA classifiers, both based on linear SVM classifiers, under bounded optimum S 2N gene selection over 200 Montecarlo 4:1 train-test partitions
  2. a The best C expressed as median: lower quartile-upper quartile.