Table 1. Summary statistics reporting the percentage of documents in which a particular field was available, and the proportion of papers labeled as Relevant and Not Relevant.
The statistics are reported for both training and final test sets.
| Field | Training | Final test |
|---|---|---|
| Title | 100 | 100 |
| Abstract | 87.17 | 87.67 |
| Authors | 99.44 | 99.63 |
| Journal | 100 | 100 |
| Publication Type | 100 | 100 |
| Keywords | 15.41 | 16.125 |
| MeSH terms | 97.67 | 98.25 |
| Chemicals | 93.86 | 94.13 |
| Affiliations | 80.94 | 79.125 |
| Label | ||
| Relevant | 19.81 | 20.25 |
| Not Relevant | 80.19 | 79.75 |