Skip to main content
Journal of Big Data

Table 3 ANOVA table for AUPRC across all levels of P

From: Synthesizing class labels for highly imbalanced credit card fraud detection data

Classifier

Df

Sum Sq

Mean SQ

F value

Pr(>F)

Df

Sum Sq

Mean SQ

Positive labels

Residuals

DT

20

1.0176

0.05088

63.52

< 2e−16

1029

0.8242

0.0008

RF

20

0.2861

0.014306

7.771

< 2e−16

1029

1.8943

0.001841

ET

20

0.1707

0.008534

4.782

2.99E−11

1029

1.8362

0.001784

LR

20

0.8621

0.0431

24.44

< 2e−16

1029

1.8149

0.00176

NB

20

0.0288

0.001441

3.251

1.96E−06

1029

0.4561

0.000443

MLP

20

0.9528

0.04764

27.41

< 2e−16

1029

1.7883

0.00174

Back to article page