Skip to main content

Table 1 Analysis of the prediction performance of individual features

From: Computational prediction of the human-microbial oral interactome

Feature AUC CA F1-score Precision Recall
+ Literature 0.781 0.722 0.723 0.721 0.726
+ Sequence 0.877 0.784 0.790 0.768 0.813
+ GO 0.817 0.742 0.748 0.735 0.760
+ COGs 0.663 0.652 0.537 0.806 0.402
+ DDIs 0.620 0.617 0.424 0.861 0.281
Final Model 0.926 0.850 0.851 0.848 0.854
  1. For each line the metrics are obtained by considering only that cluster of features on the classifier. AUC, area under the receiver operating characteristic (ROC) curve; CA, classification accuracy.