Table 5 PICO recognition performance in terms of a token-wise evaluation and a document-level filtered bag-of-words (BOW)
From: Improving reference prioritisation with PICO recognition
| Â | Token-wise | Document-level BOW | ||||
|---|---|---|---|---|---|---|
| Â | Precision | Recall | F-1 | Precision | Recall | F-1 |
Participants | 0.81 | 0.62 | 0.70 | 0.86 | 0.71 | 0.78 |
Interventions | 0.69 | 0.47 | 0.56 | 0.83 | 0.52 | 0.64 |
Outcomes | 0.66 | 0.75 | 0.70 | 0.73 | 0.81 | 0.77 |