Skip to main content

Table 1 Comparison of different gene selection methods based on accuracy

From: Identification of disease-causing genes using microarray data mining and Gene Ontology

Methods

dataset

M1

M2

M3

M4

M5

M6

M7

M8

M9

M10

Colon

86.1

86.8

89.9

91.6

89.7

91.8

93.3

91.2

93.2

94.7

DLBCL

89.2

89.6

91.9

93.7

92.5

94.4

95.8

93.6

95.4

96.8

Prostate

80.7

91.1

92.8

93.1

92.2

93.8

94.2

93.8

95.1

95.9

  1. M1 = No-sel: classification without gene selection
  2. M2 = Fisher: classification after using Fisher criteria for gene selection
  3. M3 = Fisher-R: classification after using Fisher criteria and redundancy reduction greedy approach for gene selection
  4. M4 = Fisher-RG: classification after using Fisher criteria and redundancy reduction greedy approach considering Gene Ontology information for gene selection
  5. M5 = SVMRFE: classification after using SVMRFE algorithm
  6. M6 = SVMRFE-R: classification after using SVMRFE algorithm and redundancy reduction greedy approach for gene selection
  7. M7 = SVMRFE-RG: classification after using SVMRFE algorithm and redundancy reduction greedy approach considering Gene Ontology information for gene selection
  8. M8 = Fisher-SVMRFE: classification after using combination of Fisher criteria and SVMRFE algorithm for gene selection
  9. M9 = Fisher-R-SVMRFE: classification after using proposed framework without considering Gene Ontology for gene selection
  10. M10 = Fisher-RG-SVMRFE: classification after using proposed framework for gene selection