Table 1 AUC and partial AUCs obtained by ten-fold cross-validation.

From: In vitro and in silico prediction of antibacterial interaction between essential oils via graph embedding approach

Classification

Metric

Method

Graph embedding

Classification-based

Synergistic-versus-rest

AUC

0.615 ± 0.020

0.556 ± 0.040

AUC0.5

0.211 ± 0.016

0.164 ± 0.023

AUC0.2

0.048 ± 0.004

0.033 ± 0.006

Antagonistic-versus-rest

AUC

0.576 ± 0.014

0.550 ± 0.033

AUC0.5

0.150 ± 0.010

0.159 ± 0.017

AUC0.2

0.020 ± 0.004

0.024 ± 0.003

  1. Values are means ± SD of 10 iterations, and the significantly better results are highlighted in bold (paired t-test, p < 0.01).
  2. AUC areas under the receiver operating characteristic (ROC) curve.
Back to article page