Skip to main content

Table 1 Micro- and macro-averaged performance measures across 19 experiments using random forests

From: Machine Learning for detection of viral sequences in human metagenomic datasets

Method Class Precision Recall F1-score
Micro-average Non virus 0.97 1.00 0.99
Micro-average Virus 0.92 0.05 0.10
Macro-average Non virus 0.96 1.0 0.98
Macro-average Virus 0.53 0.04 0.08
  1. The results presented are from the best model according to area under ROC curve (ROCmicro=0.789). This model used 5000 trees, balanced class weights and no down nor upsampling