Evaluating machine learning algorithms to Predict 30-day Unplanned REadmission (PURE) in Urology patients

Table 2 Evaluation of performance of regression and classification algorithms

Algorithm	Accuracy	Sensitivity	Specificity	AUC	PPV	NPV
Decision Tree	0.68	0.80	0.57	0.75	0.94	0.24
Bagged Trees	0.70	0.83	0.57	0.79	0.94	0.27
Boosted Trees	0.74	0.85	0.63	0.82	0.95	0.31
XG Boosted Trees	0.72	0.84	0.60	0.81	0.95	0.30
RandomForest	0.72	0.88	0.55	0.71	0.95	0.34
Logistic regression	0.71	0.88	0.53	0.79	0.94	0.32
LASSO regression	0.62	0.96	0.28	0.62	0.92	0.45
RIDGE regression	0.62	0.97	0.27	0.62	0.92	0.50

ISSN: 1472-6947