Table 1 Synthetic data: normally distributed: averaged AUC values achieved by different methods.

From: Edge and modular significance assessment in individual-specific networks

Method

Median AUC

Mean AUC

MultiLOO-ISN

0.628

0.632

LOO-ISN

0.582

0.591

SSN-m

0.584

0.601

KNN log(N),P

0.646

0.657

KNN 5,\(\sqrt{N}\)

0.649

0.659

Optics 5

0.608

0.595

Optics \(\sqrt{N}\)

0.532

0.529

OTS euclidean

0.628

0.637

OTS cosine

0.812

0.773

mOTS cosine

0.880

0.866

mOTS euc

0.629

0.640

mOTS glob

0.820

0.824

Spoutlier-l

0.628

0.639

Cook’s max

0.903

0.853

Cook’s med

0.920

0.859

  1. Bold values indicate the top performer of each column.
  2. The Cook’s med method achieves the best Median AUC value, while the mOTS cosine method achieves the best Mean AUC value.
Back to article page