Table 3 Predictive performance indicators of the five best models and Chengdu First People’s Hospital.

From: Development and assessment of novel machine learning models to predict the probability of postoperative nausea and vomiting for patient-controlled analgesia

Model

AUC

Accuracy

Precision

Recall

F1 Score

Specificity

SMOTE_BOR_MLP

0.5995

0.2759

0.2759

1

0.4325

0

BSMOTE_BOR_SGD

0.6501

0.6101

0.5813

0.7876

0.6689

0.4326

BSMOTE_LA_LR

0.9107

0.8342

0.7972

0.8964

0.8439

0.7720

BSMOTE_LA_XGB

0.9444

0.8795

0.9127

0.8394

0.8745

0.9197

BSMOTE_LA_CB

0.9469

0.8575

0.8943

0.8108

0.8505

0.9041

FIRST _HOSPITAL

0.8211

0.7230

0.4098

0.7618

0.5325

0.7128

  1. SMOTE synthetic minority oversampling technique, BOR boruta screening, MLP multilayer perceptron, BSMOTE borderline synthetic minority oversampling technique, SGD stochastic gradient descent, LA lasso screening, LR logistic regression, XGB extreme gradient boosting, CB category boosting, AUC area under the curve.
Back to article page