Skip to main content

Table 8 Average ranking position for AUC, AUPR and running times for all comparison methods across benchmark and generic data sets

From: LPbyCD: a new scalable and interpretable approach for Link Prediction via Community Detection in bipartite networks

 

LPbyCD

DTIP_MDHN

DTiGEMS+

ALADIN

BLM-NII

WNN-GIP

NetLapRLS

1a

2b

AUC

7.0

5.0

1.0

2.9

2.7

4.3

6.7

4.6

AUPR

6.9

5.6

1.0

2.0

3.0

6.6

6.4

4.1

tc, cpu

4.7

1.1

3.4

2.1

4.0

5.9

7.3

7.4

tc, exact

3.4

2.3

1.7

3.9

7.3

4.7

6.3

6.4

  1. a Spectral partitioning with JC\(_{NC}\) measure
  2. b The Louvain algorithm with JC\(_{NC}\) measure
  3. cRunning time