Skip to main content

Table 1 Results of the C4.5 on the missed abortion and diabetes dataset after sampled

From: A hybrid sampling algorithm combining synthetic minority over-sampling technique and edited nearest neighbor for missed abortion diagnosis

Dataset

Algorithm

Original

SM

BSM

ASM

GSM

KSM

CSM

ENN

TL

IHU

RBU

SMTOM

SMENN

Missed abortion

Precision

76.0

71.0

76.0

71.2

84.2

84.7

83.7

81.8

76.0

91.2

58.2

75.2

97.1

Sensitivity

46.4

64.1

70.7

61.9

77.9

79.5

76.7

57.1

46.4

98.2

50.0

64.7

89.8

Specificity

90.8

82.3

80.7

79.1

89.6

89.2

89.6

94.1

90.8

82.1

66.1

83.9

100.0

F-measure

75.5

73.0

75.6

70.3

83.7

84.3

83.1

80.5

75.5

90.1

57.8

74.1

96.9

MCC

42.4

47.2

51.7

41.7

67.9

69.0

66.8

57.3

42.4

81.4

16.3

49.5

92.7

AUC

70.4

74.1

74.8

73.5

86.3

86.5

86.7

77.6

70.4

90.5

59.3

74.6

91.2

Maj/Min

249/112

249/248

249/249

249/247

249/249

249/249

249/249

219/112

249/112

112/112

112/112

249/249

114/49

Diabetes

Precision

73.5

77.9

76.8

74.9

75.0

79.8

77.8

84.6

78.5

86.1

74.8

78.3

93.0

Sensitivity

59.7

82.4

83.6

81.2

71.8

82.0

79.8

83.2

70.9

84.0

75.4

82.6

95.0

Specificity

81.4

73.0

68.8

67.8

78.0

77.4

75.8

85.8

83.1

88.1

74.3

73.5

90.4

F-measure

73.6

77.7

76.1

74.4

74.9

79.7

77.8

84.5

78.5

86.0

74.8

78.0

93.0

MCC

41.7

55.6

53.0

49.4

49.9

59.5

55.6

68.9

54.2

72.1

49.6

56.3

85.7

AUC

75.1

79.6

76.1

78.9

79.0

84.8

80.7

85.4

80.4

86.6

74.2

80.2

93.4

Maj/Min

500/268

500/500

500/500

500/500

500/500

500/500

500/500

240/268

445/268

268/268

268/268

476/476

229/298