Skip to main content

Advances, Systems and Applications

Table 5 The classification results (%) over the methods

From: Ensuring security in edge computing through effective blockchain node detection

Method

Dataset

\(D_1\)

\(D_2\)

\(D_3\)

 

Metric

Pre.

Recall

F1

Pre.

Recall

F1

Pre.

Recall

F1

Random walk

DeepWalk

71.72

69.52

70.60

71.73

73.74

72.73

71.49

73.55

72.51

 

LINE

81.89

86.28

84.03

83.42

87.11

85.22

82.59

86.91

84.69

 

Node2Vec

81.31

81.86

81.58

86.85

81.98

84.34

87.52

81.96

84.65

Deep learning

GCN

76.21

76.21

76.22

77.75

73.64

75.65

77.72

74.18

75.91

 

GAT

83.31

86.58

84.92

84.76

86.69

85.71

84.83

87.23

86.01

 

GraphSAGE

83.28

87.12

85.16

84.45

88.00

86.19

89.21

88.38

86.52

Blockchain method

trans2vec

88.09

84.47

86.25

88.6

85.19

86.86

89.21

85.79

87.47

Ours

\(\textrm{T}^{2}\)A2vec

87.26

89.33

88.27

93.21

92.96

93.08

90.45

91.23

90.83