Skip to main content

Table 3 ROC scores

From: Phylogenetic tree information aids supervised learning for predicting protein-protein interaction based on distance matrices

Classification Data Representation ROC
Unsupervised (Pearson CC) MirrorTree 0.6587    
  TreeSec 0.6731    
Gaussian Kernel SVM MirrorTree 0.7067    
  TreeSec 0.7436    
  TreeSec × 10 0.8446    
  TreeSec × 10 (no NJ) 0.7196    
  TreeSec × 10 (random tree) 0.5368    
  Polynomial degree 2 3 4 5
Polynomial Kernel SVM MirrorTree 0.7212 0.5353 0.7003 0.5577
  TreeSec 0.7196 0.5657 0.7179 0.6074
  TreeSec × 10 0.7051 0.6426 0.7468 0.6619
  TreeSec × 10 (no NJ) 0.6987 0.5865 0.6907 0.6554
  TreeSec × 10 (random tree) 0.4696 0.4583 0.4503 0.5048