BMC Medical Informatics and Decision Making

Table 7 Metrics for each class for test datasets

From: An ensemble model for predicting dispositions of emergency department patients

Model	Disposition	Accuracy	AUROC	F1	Precision	Recall
BOW	Admission	0.908	0.938	0.920	0.932	0.909
	Expire	0.738	0.940	1.000	1.000	1.000
	Discharge	0.958	0.938	0.970	0.940	1.000
TF-IDF	Admission	0.910	0.942	0.920	0.931	0.910
	Expire	0.691	0.957	1.000	1.000	1.000
	Discharge	0.957	0.942	0.970	0.942	1.000

Notes BOW means Bag-of-Words, TF-IDF means term frequency–inverse document frequency, AUROC means area under the receiver operating characteristic, F1 means F1 score

Back to article page

ISSN: 1472-6947

Contact us

General enquiries: journalsubmissions@springernature.com