Skip to main content

Table 6 DMEPOS mean and standard deviation of AUPRC with varying levels of RUS (10 iterations of fivefold cross-validation)

From: Evaluating classifier performance with highly imbalanced Big Data

Class ratio

LGB

CB-16

ET

XGB-24

RF-GPU-32

1:1

0.06233

0.08935

0.92987

0.21863

0.32763

(0.00319)

(0.00453)

(0.00267)

(0.00823)

(0.01293)

1:3

0.07842

0.13984

0.93779

0.49056

0.47900

(0.00294)

(0.00422)

(0.00258)

(0.00875)

(0.01033)

1:9

0.09192

0.17826

0.94208

0.74880

0.60368

(0.00305)

(0.00497)

(0.00265)

(0.00968)

(0.00678)

1:27

0.09788

0.19499

0.94493

0.87928

0.72180

(0.00404)

(0.00510)

(0.00267)

(0.00534)

(0.00474)

1:81

0.08387

0.19489

0.94710

0.91788

0.80715

(0.00566)

(0.00508)

(0.00301)

(0.00324)

(0.00387)

Unchanged

0.07531

0.17815

0.94866

0.92811

0.83319

(0.00752)

(0.00474)

(0.00213)

(0.00316)

(0.00547)

  1. Standard deviations are below AUPRC scores in parenthesis