Table 1.

Performance characteristics of ML techniques on COVID-19 symptoms.

S. no	Model	Accuracy	Recall	Precision	F1 score	Kappa
1	Logistic regression	0.7391	0.503	0.7536	0.7195	0.5995
2	AdaBoost classifier	0.7324	0.549	0.7433	0.7093	0.5908
3	CatBoost classifier	0.7166	0.601	0.7159	0.7136	0.5817
4	Light gradient boosting machine	0.7041	0.557	0.7031	0.6997	0.561
5	Gradient boosting classifier	0.6968	0.483	0.7052	0.6816	0.537
6	Extreme gradient boosting	0.6935	0.473	0.7037	0.6757	0.5303
7	Extra trees classifier	0.6928	0.562	0.6929	0.6908	0.5494
8	Decision tree classifier	0.6909	0.59	0.697	0.6922	0.5501
9	Random forest classifier	0.6909	0.558	0.6898	0.6884	0.5459
10	SVM-linear kernel	0.6733	0.449	0.703	0.639	0.4971
11	K-neighbor classifier	0.6534	0.495	0.6474	0.6461	0.485
12	Ridge classifier	0.6487	0.345	0.4885	0.5572	0.4365
13	Quadratic discriminant analysis	0.5182	0.426	0.5352	0.5067	0.3164
14	Naive Bayes	0.4943	0.493	0.6474	0.5279	0.3152