Table 4 Boston Model Performances.
From: Personalized antibiograms for machine learning driven antibiotic selection
Antibiotic selection | Best model class | Prevalence | Average precision | AUROC |
|---|---|---|---|---|
Trime/Sulf | Gradient Boosted Tree | 0.80 | 0.85 [0.84, 0.87] | 0.60 [0.58, 0.62] |
Nitrofurantoin | Gradient Boosted Tree | 0.89 | 0.91 [0.90, 0.92] | 0.57 [0.54, 0.61] |
Ciprofloxacin | Lasso | 0.94 | 0.95 [0.95, 0.96] | 0.64 [0.60, 0.68] |
Levofloxacin | Lasso | 0.94 | 0.96 [0.95, 0.96] | 0.64 [0.60, 0.67] |
