Skip to main content

Table 5 Performance comparison of feature selection frameworks

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

 

GA

SFS

Classifier

mean ± std

#genes

mean ± std

#genes

LDA

99.682 ± 0.12

9.33

92.282 ± 3.22

2.5

SVM

99.082 ± 0.25

15.67

95.185 ± 2.36

4

NaiveBayes

97.847 ± 0.16

12.83

93.156 ± 3.11

3.67

C-MANTEC

98.150 ± 0.25

9.83

92.960 ± 3.46

2.5

kNN

98.688 ± 0.14

15.17

95.249 ± 2.36

4.33

MLP

99.798 ± 0.08

9

93.401 ± 3.08

2.5

  1. Average performance comparison among two different feature selection frameworks (GA and SFS) and six classifiers (LDA, SVM, NaiveBayes, C-MANTEC, kNN and MLP) over all dataset.
Back to article page

Theoretical Biology and Medical Modelling

ISSN: 1742-4682