Skip to main content

Table 4 The comparison of applied strategies for facing imbalanced dataset challenge

From: DrugRep-HeSiaGraph: when heterogenous siamese neural network meets knowledge graphs for drug repurposing

Strategy

ACC

AUC-ROC

AUC-PR

BS

MCC

F1-score

IMB

80.64

92.69

91.69

12.94

62.76

84.21

RUS

83.96

92.48

91.56

8.25

68.07

84.14

ROS

85.72

92.61

90.76

10.92

71.61

82.67

OSMOTE

84.62

92.19

90.49

11.18

69.34

82.07

SORU

86.38

92.57

91.55

10.69

73.02

84.23