Table 1.

Test set performance for the three tasks

	True positive	False negative	False positive	Recall	Precision	F-score
Task 1: Concepts Task
System 1.1	37 646	7363	5683	0.8364	0.8688	0.8523
System 1.2	36 776	8233	6125	0.8170	0.8572	0.8366
System 1.3	37 663	7346	5787	0.8367	0.8668	0.8515
Task 2: Assertions Task
System 2.1	17 366	1184	1184	0.9362	0.9362	0.9362
System 2.2	17 338	1212	1212	0.9347	0.9347	0.9347
System 2.3	17 197	1353	1353	0.9271	0.9271	0.9271
Task 3: Relations Task
System 3.1	6296	2809	1965	0.6902	0.7611	0.7239
System 3.2	6269	2801	1896	0.6911	0.7677	0.7274
System 3.3	6288	2782	1838	0.6932	0.7738	0.7313