Table 1.

Performance of different natural language processing systems on the evaluation set under 4 conditions using 100, 200, 500, and 1000 target-domain training examples^a.

System		AUC-ROCb				Average precision
		100	200	500	1000	100	200	500	1000
SourceOnly		0.739	0.739	0.739	0.739	0.811	0.811	0.811	0.811
TargetOnly		0.728	0.749	0.769	0.782	0.799	0.816	0.833	0.844
ADS-fsac		0.746	0.756	0.776	0.790	0.815	0.823	0.839	0.850
ADS-sdsd		0.751	0.759	0.775	0.786	0.819	0.826	0.838	0.847
ADS-fsa vs ADS-sdse
	t99	4.25	2.79		8.78	3.81	3.04		11.58
	P values	<.001	.01		<.001	<.001	.003		<.001

^aThe highest performance scores are italicized.

^bAUC-ROC: area under the receiver operating characteristic curve.

^cADS-fsa: adapted distant supervision-feature space augmentation.

^dADS-sds: adapted distant supervision-supervised distant supervision.

^eThe P values for difference between ADS-fsa and SourceOnly, ADS-sds and SourceOnly, ADS-fsa and TargetOnly, and ADS-sds and TargetOnly are <.001 (t₉₉ ranges from 4.84 to 133.31) for all metrics under all conditions. We report the P values (if the P value ≤.05) and the corresponding t₉₉ values for difference between ADS-fsa and ADS-sds.