Table 2.
Performance (in %) of the baseline relation extraction system and the incremental contribution of different system modules, on the test set of the ADE corpus
| System | Precision | Recall | F-score |
|---|---|---|---|
| Baseline |
8.9 |
78.4 |
16.1 |
| + NLP module |
21.1 |
82.9 |
33.6 |
| + Knowledge base |
32.8 |
78.1 |
46.2 |
| + Relation-type filtering | 38.1 | 74.8 | 50.5 |