Skip to main content

Table 3 The performance comparison of robustness and generalization of multiple models and the independent test results of ATT-MLP are presented. The MCC and AUC are for Matthews correlation coefficient and Area Under the ROC curve in the table

From: Effective attention-based network for syndrome differentiation of AIDS

Methods

Syndromes

Naive Bayes

 

SVM

 

RF

 

MLP

 

ATT-MLP

  
 

MCC

AUC

MCC

AUC

MCC

AUC

MCC

AUC

MCC

AUC

P

S1

0.41

0.63

0.34

0.69

0.58

0.74

0.50

0.71

0.67

0.79

<.001

S2

0.53

0.70

0.38

0.72

0.65

0.74

0.50

0.72

0.69

0.79

<.001

S3

0.34

0.64

0.32

0.68

0.63

0.75

0.47

0.73

0.62

0.74

<.001

S4

0.55

0.76

0.40

0.73

0.68

0.78

0.61

0.74

0.71

0.82

<.001

S5

0.28

0.60

0.37

0.72

0.60

0.73

0.27

0.70

0.68

0.78

<.001

S6

0.49

0.68

0.44

0.75

0.67

0.80

0.40

0.74

0.63

0.76

<.001

S7

0.51

0.71

0.47

0.74

0.56

0.83

0.51

0.76

0.73

0.84

<.001

Avg.

0.44

0.68

0.39

0.72

0.62

0.77

0.47

0.73

0.67

0.79

Â