Skip to main content

Table 14 Average F-measure of different state-of-the-art under-sampling and over-sampling methods as compared with the proposed approach

From: A multi-manifold learning based instance weighting and under-sampling for imbalanced data classification problems

Name

BIDC1 [51]

BIDC2 [51]

K-US [50]

NB-Rec [49]

DB_US [48]

Proposed

ecoli2

0.81

0.83

0.59

0.91

0.96

0.90 ± 0.142

ecoli3

NA

NA

0.71

0.63

0.71

0.86 ± 0.195

glass0

0.63

0.66

0.61

0.61

0.61

0.79 ± 0.116

new-thyroid1

NA

NA

0.73

0.92

1.00

0.89 ± 0.206

Pima

NA

NA

0.73

0.58

0.71

0.63 ± 0.086

segment0

1.00

1.00

0.83

1.00

1.00

0.98 ± 0.017

wisconsin

NA

NA

0.97

0.97

0.97

0.95 ± 0.053

Average F-measure

0.81(4)

0.83(3)

0.74(6)

0.80(5)

0.85(2)

0.86(1)