Table 8.
Evaluation results for the Chemical Exposure Assessment (CEA) text classification task
| Document classification | Sentence classification | |||||
|---|---|---|---|---|---|---|
| Model | Precision | Recall | F1 | Precision | Recall | F1 |
| Baseline (no retrofitting) | 89.5 | 87.1 | 88.3 | 66.2 | 62.8 | 64.5 |
| 22-classes retrofitted | 89.9 | 87.5 | 88.7* | 67.3 | 62.1 | 64.6 |
| 117-subclasses retrofitted | 89.2 | 88.6 | 88.9* | 66.3 | 60.3 | 63.2* |
Baseline model is a skip-gram model without any retrofitting. All figures are micro-averages expressed as percentages (Bold denotes the best F1-score, * denotes statistically significant scores with respect to the baseline)