TABLE 3.

Performances (SVM and LR) of two active learning methods with single machine learning (ML) algorithms.

Two AL methods’ performances for SVM		TP	FP	TN	FN	Recall	Precision	F-score	Precision (recall = 0.99)
Uncertainty sampling, a single ML algorithm, and no validation data update	1st round	190	2	198	10	0.95	0.99	0.97	0.97
	2nd round	134	0	200	66	0.67	1.00	0.80	0.94
Uncertainty sampling and random negative sampling, a single ML algorithm, and no validation data update	1st round	176	0	400	24	0.88	1.00	0.94	0.96
	2nd round	138	0	400	62	0.69	1.00	0.82	0.94
Two AL methods’ performances for LG		TP	FP	TN	FN	Recall	Precision	F-score	Precision (Recall = 0.99)
Uncertainty sampling, a single ML algorithm, and no validation data update	1st round	187	7	193	13	0.94	0.97	0.95	0.96
	2nd round	161	1	199	39	0.81	0.99	0.89	0.94
Uncertainty sampling and random negative sampling, a single ML algorithm, and no validation data update	1st round	176	5	395	24	0.88	0.97	0.92	0.95
	2nd round	159	1	399	41	0.80	0.99	0.88	0.94