Skip to main content

Table 3 Binary classification results with traditional machine learning and deep learning methods

From: Development of models for classification of action between heat-clearing herbs and blood-activating stasis-resolving herbs based on theory of traditional Chinese medicine

Data set

Type of models

Sensitivity (%)

Specificity (%)

Precision (%)

Accuracy (%)

Calibration set

SVM

94.4

72.4

89.3

88

DBN

/

/

/

/

kNN

91.5

75.9

90.3

87

CNN

/

/

/

/

Validation set

LS-SVM

82.4

81.3

82.4

81.8

DBN

100

100

100

100

kNN

82.4

62.5

70.0

72.7

CNN

100

100

100

100

External validation set

SVM

93.3

75.0

87.5

85.0

DBN

100

100

100

100

kNN

86.7

80.0

92.9

85.0

CNN

100

100

100

100