Table 2 20-fold cross validation performance estimation on KEGG and KEGG+NetPath.

From: A simple feature construction method for predicting upstream/downstream signal flow in human protein-protein interaction networks

 

KEGG

KEGG+NetPath-EC

SP

SE

MCC

SP

SE

MCC

Positive

0.7560

0.7560

0.5920

0.7782

0.7835

0.6277

Negative

0.7560

0.7560

0.5920

0.7820

0.7767

0.6270

[Acc; MCC]

[75.60%; 0.5920]

  

[78.01%; 0.7162]

  

[ROC-AUC]

[0.8399]

  

[0.8394]

  

F1 Score

0.7560

  

0.7808

  

Macro_accuracy

74.12%

  

75.87%

  
  1. Macro_accuracy is used to measure the reliability that the proposed method predicts upstream/downstream relations and the other metrics are used to estimate the instance-level performance.
Back to article page