Predicting risk of sepsis, comparison between machine learning methods: a case study of a Virginia hospital

Table 2 Results of six proposed machine learning methods

	Logistic Regression	Boosted Tree	Bootstrap Forest	CART	SVM	Naïve Bayes
Accuracy	0.933	0.936	0.935	0.933	0.937	0.899
Specificity	0.984	0.983	0.984	0.986	0.983	0.923
precision	0.607	0.634	0.620	0.639	0.641	0.385
Recall	0.297	0.349	0.323	0.282	0.362	0.385
F-1 score	0.399	0.450	0.425	0.391	0.463	0.468
Misclassification rate	0.066	0.063	0.064	0.066	0.063	0.100
AUC for training data	0.887	0.901	0.908	0.901	0.829	0.884
R-square	0.327	0.355	0.366	0.354	0.223	0.033

ISSN: 2047-783X