Table 2.

Performance metrics by age, gender, region and language using XGBoost

Data grouping	CA	Precision	Recall	F1	AUC
All	0.80	0.79	0.80	0.79	0.84
By age, years
18–24	0.68	0.68	0.68	0.68	0.74
25–34	0.71	0.70	0.71	0.70	0.75
35–44	0.77	0.74	0.77	0.75	0.75
45–54	0.82	0.79	0.82	0.79	0.78
55–64	0.86	0.83	0.86	0.83	0.79
65–74	0.91	0.88	0.91	0.88	0.76
75–84	0.94	0.91	0.94	0.92	0.70
By gender
Male	0.82	0.80	0.82	0.80	0.81
Female	0.79	0.77	0.79	0.78	0.83
By region
Core Anglosphere	0.79	0.79	0.79	0.79	0.84
Latin America	0.82	0.80	0.82	0.80	0.82
Europe	0.85	0.84	0.85	0.84	0.84
Middle East	0.74	0.72	0.74	0.72	0.72
North Africa	0.81	0.78	0.81	0.78	0.72
West Africa	0.82	0.79	0.82	0.80	0.72
Sub-Saharan Africa	0.75	0.74	0.75	0.74	0.79
South Asia	0.75	0.75	0.75	0.75	0.81
South-East Asia	0.77	0.76	0.77	0.77	0.82
By language
English	0.78	0.77	0.78	0.78	0.83
Spanish	0.83	0.81	0.83	0.81	0.84
French	0.86	0.83	0.86	0.83	0.72
Arabic	0.76	0.73	0.76	0.74	0.73

AUC, area under the ROC curve; CA, classification accuracy.