Skip to main content

Table 5 Results of the ablation experiments on model DAEGCNDF based on 10-fold cross-validation

From: Predicting potential microbe-disease associations based on auto-encoder and graph convolution network

Dataset:

MDAID

Experiments

Acc(%)

Pre (%)

Recall (%)

F1-score (%)

AUC (%)

AUPR (%)

LRF

90.75

90.75

90.74

90.74

96.85

96.78

HRF

86.67

86.67

86.65

86.65

94.63

94.64

LHRF

90.89

90.89

90.88

90.87

97.00

96.90

Dataset:

HMDAD

Experiments

Acc(%)

Pre (%)

Recall (%)

F1-score (%)

AUC (%)

AUPR (%)

LRF

85.56

85.86

85.77

85.56

95.89

96.11

HRF

85.89

85.48

85.56

85.51

95.04

96.12

LHRF

88.11

88.35

88.21

88.08

96.32

96.72

  1. The bold result indicates the best one in each column