Skip to main content

Table 1 AUC results for graph-level anomaly detection

From: Recognition of Yuan blue and white porcelain produced in Jingdezhen based on graph anomaly detection combining portable X-ray fluorescence spectrometry

AUC

EEXPT.1

EEXPT.2

EEXPT.3

MEAN

STD

OCSVM

82.69%

54.41%

54.32%

63.81%

16.35%

IF

91.83%

82.37%

81.09%

85.10%

5.87%

LOF

92.63%

92.31%

93.43%

92.79%

0.58%

GAAN

75.16%

53.69%

58.81%

62.55%

11.21%

OCGNN

89.58%

91.19%

90.38%

90.38%

0.81%

GCNAE

92.63%

92.63%

92.63%

92.63%

0.00%

DOMINANT

86.70%

88.30%

86.70%

87.23%

0.92%

CONAD

90.71%

90.06%

90.06%

90.28%

0.38%

GUIDE

92.63%

92.79%

92.95%

92.79%

0.16%

GRAM

95.19%

94.87%

95.35%

95.14%

0.24%

  1. Bold values indicate that GRAM achieves the best results compared to other methods