Table 1-.
Precision, recall, F1-measure, Stanford NLP
| Class | Metric | Dataset |
||
|---|---|---|---|---|
| Baseline | HCR | HPV | ||
| Positive | Precision | 54% | 11% | 7.3% |
| Recall | 73% | 62.7% | 47% | |
| F1-measure | 62% | 18.8% | 12.7% | |
| Negative | Precision | 39% | 47.2% | 58.2% |
| Recall | 71% | 46.5% | 35.8% | |
| F1-measure | 51% | 46.8% | 44.3% | |
| Neutral | Precision | 74% | 38.6% | 39% |
| Recall | 34% | 20% | 34% | |
| F1-measure | 47% | 26.3% | 36.6% | |