Figure 1
From: Pooling annotated corpora for clinical concept extraction
Performance measures of the taggers. The plots A, B and C show the F1-score, precision and recall respectively. Each line in the figure corresponds to the test set (MCR or i2b2) and the evaluation method: Exact (E) with solid lines or Overlap (O) with dashed lines. The horizontal axis indicates the training sets: MCR, i2b2 and combined (MCR + i2b2).