Study on the semi-supervised learning-based patient similarity from heterogeneous electronic medical records

Table 2 Comparison of predictive performance between similarity-based models with other machine learning models

	Area under the receiver operating characteristic curve	F1-score	Cross-entropy loss
Naive bayes	0.92	0.80	3.60
Decision tree	0.91	0.73	0.72
Random forest	0.94	0.78	0.75
AdaBoost	0.86	0.70	1.16
R + E/L + E combination	0.89	0.67	1.84
R + L/E + L combination	0.95	0.84	1.23

ISSN: 1472-6947