Table 5.

Continents prediction statistics for Random Forest for CSD17 sampling date using proGenomes database with the top 16 features from RFE.

	CSD17
	Sensitivity	Specificity	Accuracy
Sub Saharan Africa	1.00	1.00	1.00
North America	0.90	0.99	0.95
East Asia	0.89	0.79	0.84
Europe	0.69	0.91	0.80
South America	0.45	0.996	0.73
Middle East	0.17	1.00	0.58

The overall accuracy is 0.81 (0.76, 0.86). The data are sorted by balanced accuracy which is the average between sensitivity and specificity.