Skip to main content

Table 3 Results of the proposed method compared to generic classifiers [21, 22, 24, 25]

From: Justified granulation aided noninvasive liver fibrosis classification system

Method

Overall accuracy [%]

Specificity [%]

Sensitivity [%]

Ŝ 1

Ŝ 2

Ŝ 3

Ŝ 1

Ŝ 2

Ŝ 3

Granular model

67.9

62.1

93.1

91.8

86.8

40.2

74.6

Random Tree

61.4

71.4

76.6

90.5

66.7

57.8

55.9

PNN

60.7

70.8

88.8

80.1

76.7

32.4

74.6

Naïve Bayes

59.7

75.8

63.3

96.1

62.0

62.7

49.2

Auto-Binned IBk

58.6

54.0

79.3

97.0

77.5

38.2

52.5

RecBF-DDA

56.9

60.4

72.8

96.2

65.2

53.1

46.6

IBk (k-NN)

49.3

30.4

82.4

99.1

87.6

20.6

15.3