Table 3.

Results from different ML algorithms.

Measures	Subset	AUC	Accuracy	F1	Recall	Precision
Time series	Train	0.49	0.51	0.00	0.27	0.35
	Test	0.50	0.51	0.34	0.50	0.26
PC	Train	0.67	0.67	0.69	0.67	0.66
	Test	0.60	0.60	0.60	0.60	0.60
SC	Train	0.98	0.98	0.97	0.98	0.98
	Test	0.98	0.98	0.98	0.98	0.98
GC	Train	0.51	0.52	0.00	0.32	0.37
	Test	0.50	0.51	0.34	0.50	0.26
BM	Train	0.75	0.75	0.72	0.75	0.75
	Test	0.75	0.75	0.75	0.75	0.75
SCC	Train	0.67	0.67	0.65	0.67	0.66
	Test	0.62	0.62	0.62	0.62	0.62
GL	Train	0.66	0.66	0.65	0.66	0.66
	Test	0.57	0.58	0.57	0.57	0.57
LW	Train	0.66	0.66	0.64	0.66	0.65
	Test	0.58	0.58	0.58	0.58	0.58
MI	Train	0.49	0.50	0.40	0.50	0.49
	Test	0.49	0.49	0.49	0.49	0.49
TE	Train	0.90	0.90	0.89	0.90	0.90
	Test	0.91	0.91	0.91	0.91	0.91

The best MLs were RF and LR, whose performances are highlighted.

Significant values are in bold.