Application of machine learning models based on decision trees in classifying the factors affecting mortality of COVID-19 patients in Hamadan, Iran

Table 2 Assessing the accuracy of decision tree models in classifying the COVID-19 death

Model	Subset	Confusion matrix				Evaluation metric
Model	Subset	TP	FP	FN	TN	Recall	F1-score	Accuracy
LMT	Train	1225	299	73	133	0.9437	0.8681	0.7850
	Test	514	130	41	55	0.9261	0.8573	0.7689
	Total	1739	429	113	189	0.9383	0.8650	0.7804
C4.5	Train	1217	284	81	148	0.9376	0.8696	0.7891
	Test	510	120	45	65	0.9189	0.8608	0.7770
	Total	1728	412	125	205	0.9325	0.8655	0.7826
C5.0	Train	1192	276	106	156	0.9183	0.8619	0.7792
	Test	514	130	41	55	0.9261	0.8574	0.7689
	Total	1739	429	114	188	0.9384	0.8649	0.7802
CART	Train	1217	284	81	148	0.9376	0.8696	0.7891
	Test	530	136	25	49	0.9550	0.8681	0.7824
	Total	1739	421	114	196	0.9385	0.8667	0.7834

ISSN: 1472-6947