Skip to main content

Table 3 Comparison of prediction accuracy of ELOHS using tenfold cross-validation for ELOHS with all the features

From: Preadmission assessment of extended length of hospital stay with RFECV-ETC and hospital-specific data

Algorithm

Recall

Precision

F1-score

Balanced accuracy

ROC AUC

KNN

0.807 ± 0.045

0.821 ± 0.032

0.803 ± 0.05

0.807 ± 0.045

0.892 ± 0.033

GBM

0.746 ± 0.12

0.77 ± 0.1

0.732 ± 0.136

0.746 ± 0.12

0.876 ± 0.075

DTC

0.818 ± 0.061

0.845 ± 0.054

0.814 ± 0.063

0.818 ± 0.061

0.907 ± 0.054

ADB

0.715 ± 0.104

0.732 ± 0.09

0.703 ± 0.118

0.715 ± 0.105

0.835 ± 0.078

ETC

0.885 ± 0.063

0.9 ± 0.052

0.883 ± 0.066

0.885 ± 0.063

0.952 ± 0.039

SVM

0.723 ± 0.076

0.726 ± 0.072

0.719 ± 0.083

0.723 ± 0.076

0.805 ± 0.07

XGB

0.77 ± 0.11

0.809 ± 0.083

0.755 ± 0.128

0.77 ± 0.11

0.927 ± 0.075

RF

0.859 ± 0.078

0.883 ± 0.061

0.855 ± 0.082

0.859 ± 0.078

0.953 ± 0.053