Table 3 Confusion matrices for classifiers with the best area under the receiver-operator curve (AUROC) among all patients, with the threshold set to match a 50% true positive rate

From: Validation of sleep-based actigraphy machine learning models for prediction of preterm birth

(a) Actigraphy/Gaussian naive Bayes

 

PN

PP

TN

0.605

0.258

TP

0.076

0.061

(b) Case Reports/Gaussian naive Bayes

 

PN

PP

TN

0.701

0.162

TP

0.076

0.061

(c) All/Gaussian naive Bayes

 

PN

PP

TN

0.734

0.129

TP

0.095

0.043

  1. PN is predicted negative, PP is predicted positive, TN is true negative, and TP is true positive.