Skip to main content

Table 2 The accuracy scores of Relational Graph Convolutional Networks (R-GCN), RDF2VEC and our proposed approaches on four benchmark data sets

From: MINDWALC: mining interpretable, discriminative walks for classification of nodes in a knowledge graph

data set

R-GCN

RDF2VEC

Tree

Forest

Transf+LR

Transf+RF

AIFB

95.83 ±0.62

88.88 ± 0.00

85.83 ± 2.43

90.83 ± 2.29

85.28 ± 1.34

86.39 ± 0.88

BGS

83.10 ± 0.80

87.24 ± 0.89

87.58 ± 1.78

91.38 ± 2.44

89.31 ± 1.09

93.10 ± 0

MUTAG

73.23 ± 0.48

67.20 ± 1.24

68.97 ± 4.46

74.41 ± 2.88

78.82 ± 1.86

73.24 ± 1.16

AM

89.29 ± 0.35

88.33 ± 0.61

86.81 ± 1.27

88.33 ± 0.73

90.05 ± 0.05

89.89 ± 1.14