TABLE 3.

Test set performances of kNN, RF, and representative spectrum classification models trained on real data only. Results are reported as average ± SD over ten random splits (see Materials and Method section for more details).

kNN
	Precision	Recall	F1‐score
BZ	0.75 ± 0.09	0.81 ± 0.11	0.77 ± 0.08
PD	0.85 ± 0.05	0.83 ± 0.07	0.83 ± 0.05
HR	0.84 ± 0.10	0.90 ± 0.09	0.87 ± 0.07

RF
	Precision	Recall	F1‐score
BZ	0.85 ± 0.08	0.84 ± 0.08	0.84 ± 0.07
PD	0.88 ± 0.07	0.85 ± 0.05	0.86 ± 0.05
HR	0.90 ± 0.07	0.95 ± 0.05	0.92 ± 0.05

Representative spectrum
	Precision	Recall	F1‐score
BZ	0.84 ± 0.11	0.55 ± 0.12	0.65 ± 0.10
PD	0.76 ± 0.08	0.51 ± 0.09	0.61 ± 0.08
HR	0.52 ± 0.05	0.99 ± 0.03	0.68 ± 0.04