Classifying the Mortality of People with Underlying Health Conditions Affected by COVID-19 Using Machine Learning Techniques

Table 6

Experiment results.

Model	Evaluation matrix	Exp. 1	Exp. 2	Exp. 3
SVM	Accuracy	81.385%	78.355%	77.922%
	TPR	0.814	0.784	0.779
	FPR	0.181	0214	0.220
RF	Accuracy	82.251%	74.026%	80.52%
	TPR	0.823	0.740	0.805
	FPR	0.175	0.258	0.193
J.48	Accuracy	78.355%	78.788%	76.623%
	TPR	0.784	0.788	0.766
	FPR	0.215	0.209	0.232
NB	Accuracy	78.788%	80.087%	79.654%
	TPR	0.788	0.801	0.797
	FPR	0.207	0.194	0.198
LR	Accuracy	82.251%	81.818%	79.221%
	TPR	0.823	0.818	0.792
	FPR	0.173	0.175	0.202
Bagging	Accuracy	83.55%	78.788%	83.117%
	TPR	0.835	0.788	0.831
	FPR	0.160	0.207	0.165
Threshold selector	Accuracy	80.952%	81.385%	80.087%
	TPR	0.810	0.814	0.801
	FPR	0.185	0.181	0.200