Table 2

Range accuracies for the mixed oral cavity carcinoma metastasis dataset

Range of Number of Feature Genes

Kernel

Noise

1–10

11 – 20

21 – 30

31 – 40

41 – 50

51 – 60

61 – 70

71–80

81 – 90

91 – 100


Linear

raw

73.8

79.3

77.0

73.5

72.8

70.0

70.0

70.0

70.0

68.5

SVD

66.3

76.8

74.0

74.8

75.0

76.3

77.5

77.5

77.5

77.5

Polynomial

raw

77.8

81.8

77.3

74.3

72.8

70.0

70.0

70.0

70.0

67.8

SVD

68.0

79.5

79.0

76.3

75.0

76.0

77.5

77.5

77.5

77.5

RBF

raw

77.0

82.3

78.0

75.8

73.5

68.3

68.5

70.0

68.8

66.5

SVD

68.0

78.5

79.0

81.8

80.8

80.3

79.5

77.8

78.8

79.0

EKM

raw

74.5

78.5

76.8

72.3

67.3

66.8

65.5

65.5

65.0

65.0

SVD

68.8

80.0

79.0

81.8

80.8

80.3

80.0

79.5

81.0

80.3

Saigo

raw

77.5

81.5

80.3

73.5

70.3

70.0

69.0

68.3

68.0

67.5

SVD

68.0

77.8

77.5

76.0

75.0

76.8

77.8

78.3

78.0

78.8

kNND-ME

raw

78.5

84.0

82.8

77.8

73.3

71.3

70.8

70.5

72.0

70.3

SVD

69.5

78.5

79.0

79.5

77.5

79.0

80.5

80.0

80.3

83.3


Maximum accuracy in each range of feature gene number is shown in bold.

Fujibuchi and Kato BMC Bioinformatics 2007 8:267   doi:10.1186/1471-2105-8-267

Open Data