Skip to main content

Table 6 Differences between classifiers.

From: Application of genetic algorithms and constructive neural networks for the analysis of microarray cancer data

FS procedure

Dataset

p-value

Control

Statistically different classifiers

SFS

Leukemia

< e −16

LDA

SVM

 

Lung

< e −16

LDA

kNN, NB

 

Colon

< e −16

LDA

SVM, kNN

 

Breast

< e −16

NB

kNN, SVM

 

Ovarian

< e −16

CM

LDA, NN

 

Prostate

< e −16

CM

NB, SVM

GA

Leukemia

< e −16

CM

NB, NN, LDA, SVM, kNN

 

Lung

< e −16

CM

SVM, NB, NN

 

Colon

< e −16

SVM

LDA, NN

 

Breast

< e −16

SVM

NN, LDA

 

Ovarian

< e −16

CM

SVM, NN

 

Prostate

< e −16

CM

NN, LDA

  1. Differences between classifiers for the two feature selection (FS) procedures used (first column). The lowest performance classifier is taken as control group and the last column of the table lists the classifiers that lead to statistically significant results (corresponding p-value indicated in the third column).