Skip to main content

Table 3 Average rankings according to the Friedman statistical test

From: Reduction strategies for hierarchical multi-label classification in protein function prediction

Method

\(AU(\overline {PRC})\)

Method

\(\overline {AUPRC}_{w}\)

Method

\(\overline {AUPRC}\)

Predicted

1.35

Predicted

1.25

Clus-HSC

1.90

NoLabels

2.50

NoLabels

2.75

Clus-HMC

2.20

True

3.40

True

3.45

Clus-SC

2.70

Clus-HMC

3.55

Clus-HMC

3.85

Predicted

4.10

Labels

4.90

Clus-SC

5.55

NoLabels

5.45

hmAnt-Miner

5.30

Clus-HSC

5.70

hmAnt-Miner

5.50

Clus-HSC

7.20

Labels

6.65

True

6.15

Clus-SC

7.80

hmAnt-Miner

6.80

Labels

8.00