Skip to main content

Advances, Systems and Applications

Table 6 Three metrics results-PPD

From: MF-GCN-LSTM: a cloud-edge distributed framework for key positions prediction in grid projects

Metrics

Models

 

SVM

Static GCN

LSTM

MF-GCN-LSTM

Pre

PPD0

0.53

0.67

0.91

1.00

 

PPD1

0.70

0.81

0.94

1.00

 

PPD2

0.59

0.82

0.84

1.00

 

AVG.

0.61

0.77

0.91

1.00

Rec

PPD0

0.81

0.63

0.85

0.80

 

PPD1

0.90

0.71

0.88

0.94

 

PPD2

0.66

0.78

0.75

0.84

 

AVG.

0.79

0.71

0.82

0.86

F1

PPD0

0.64

0.65

0.87

0.88

 

PPD1

0.79

0.75

0.91

0.97

 

PPD2

0.63

0.79

0.83

0.91

 

AVG.

0.68

0.73

0.87

0.92

 

Diff.

0.16

0.14

0.08

0.09

  1. The boldface in the tables represents the optimal values among the four models, and the underline represents the suboptimal values