Table 5.

Performance (in %) of a machine-learning based (jSRE) relation extraction system [[32]] and the knowledge-based system on a subset of the ADE test corpus (see text)

Training set (abstracts)	Machine learning			Knowledge base
	Precision	Recall	F-score	Precision	Recall	F-score
0	n/a	n/a	n/a	88.5	88.6	88.5
10	58	6	55	89.1	88.2	88.6
50	79	87	82	91.8	86.1	88.8