Table 2 Ratio of each sentence type at the token level based on MER and KE tags in the training data
From: Sentences, entities, and keyphrases extraction from consumer health forums using multi-task learning
Entity type | Keyphrase? | Sentence type (%) | ||
|---|---|---|---|---|
Background | Question | Ignore | ||
Disease | \(\checkmark\) | 58.41 | 40.27 | 1.33 |
\(\times\) | 75.71 | 20.00 | 4.29 | |
Symptom | \(\checkmark\) | 93.59 | 6.41 | 0.00 |
\(\times\) | 97.98 | 2.02 | 0.00 | |
Drug | \(\checkmark\) | 74.62 | 25.38 | 0.00 |
\(\times\) | 89.83 | 10.17 | 0.00 | |
Treatment | \(\checkmark\) | 63.14 | 36.83 | 0.00 |
\(\times\) | 90.11 | 9.89 | 0.00 | |