BMC Medical Informatics and Decision Making

Table 5 Performance analysis of applied machine learning methods for unseen test data

From: Enhanced interpretable thyroid disease diagnosis by leveraging synthetic oversampling and machine learning models

Technique	Accuracy	Target class	Precision	Recall	F1
LR	0.80	Negative	0.75	0.91	0.82
		Sick	0.89	0.70	0.78
		Average	0.82	0.80	0.80
LSVM	0.80	Negative	0.75	0.91	0.82
		Sick	0.89	0.70	0.78
		Average	0.82	0.80	0.80
RF	0.93	Negative	0.88	1.00	0.94
		Sick	1.00	0.87	0.93
		Average	0.94	0.93	0.93
LGBM	0.96	Negative	0.92	1.00	0.96
		Sick	1.00	0.91	0.95
		Average	0.96	0.96	0.96

Back to article page

ISSN: 1472-6947

Contact us

General enquiries: journalsubmissions@springernature.com