Skip to main content

Table 1 The accuracy of the baseline system built by randomly selecting genes from 7,070 genes in the leukemia dataset.

From: Feature selection and classification for microarray data analysis: Evolutionary methods for identifying predictive genes

Population size

Feature size

Training data [%]

Test data [%]

Max

Average

Max

Average

10

30

89.47

85.79

79.41

64.41

 

50

89.47

84.21

76.47

63.53

30

30

94.74

98.42

85.29

68.82

 

50

97.37

96.05

73.53

67.35

50

30

100.00

98.42

85.29

70.29

 

50

100.00

98.42

85.29

72.64