readme
Browse files
README.md
CHANGED
|
@@ -7,4 +7,23 @@ tags:
|
|
| 7 |
- pubmed
|
| 8 |
- ner
|
| 9 |
- pubmedbert
|
| 10 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 7 |
- pubmed
|
| 8 |
- ner
|
| 9 |
- pubmedbert
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
# PubMed PICO extraction (NER model)
|
| 13 |
+
|
| 14 |
+
This is a fine-tuned multi-class NER model for PICO labels within PubMed biomedical database abstracts, trained with [NERDA](https://ebanalyse.github.io/NERDA/).
|
| 15 |
+
|
| 16 |
+
The mode predicts:
|
| 17 |
+
|
| 18 |
+
| Class | **F1-Score** | Precision | Recall |
|
| 19 |
+
|B-Patient | **0.72** | 0.69 | 0.76
|
| 20 |
+
|I-Patient | **0.77** | 0.74 | 0.81
|
| 21 |
+
|
| 22 |
+
|B-Intervention | **0.62** | 0.61 | 0.63 |
|
| 23 |
+
|I-Intervention | **0.70** | 0.70 | 0.71 |
|
| 24 |
+
|
| 25 |
+
|B-Control | **0.59** | 0.56 | 0.63 |
|
| 26 |
+
|I-Control | **0.62** | 0.64 | 0.61 |
|
| 27 |
+
|
| 28 |
+
|B-Outcome | **0.49** | 0.50 | 0.49 |
|
| 29 |
+
|I-Outcome | **0.56** | 0.63 | 0.51 |
|