Table 4 The performance of the classification models on the non-redundant testing data set.

From: Artificial Intelligence and Machine learning based prediction of resistant and susceptible mutations in Mycobacterium tuberculosis

Gene

Measure/Methods

NB

SVM

ANN

kNN

rpoB

Accuracy

90.90%

86.36%

90.90%

95.45%

AUC

0.97

0.85

1

0.95

InhA

Accuracy

100%

60%

81.81%

100%

AUC

1

0.5

0.92

1

katG

Accuracy

98%

70%

98%

96%

AUC

0.98

0.69

1

0.97

pncA

Accuracy

93.75%

81.25%

97.91%

97.91%

AUC

0.97

0.82

1

0.98

gyrA

Accuracy

92.85%

78.57%

100%

85.71%

AUC

0.97

0.77

1

0.86

gyrB

Accuracy

66.66%

77.77%

88.88%

88.88%

AUC

0.75

0.8

1

0.92

Back to article page