E.19. Universal POS Tags
The universal tags are coarser than the language-specific tags, but enable tracking and comparison across languages.
To return universal POS tags in place of language-specific tags, use the Annotated Data Model (ADM) and BaseLinguisticsFactory to set BaseLinguisticsOption.universalPosTags and BaseLinguisticsOption.deliverExtendedTags to "true". See Returning Universal POS Tags [38].
RBL-JE can use universal POS tags from the UPT-16 set for 20 languages.
| Language | Supported |
|---|---|
| Arabic | ✓ |
| Czech | ✓ |
| Chinese | ✓ |
| Danish | |
| Dutch | ✓ |
| English | ✓ |
| Finnish | |
| French | ✓ |
| German | ✓ |
| Greek | ✓ |
| Hebrew | |
| Hungarian | ✓ |
| Italian | ✓ |
| Japanese | ✓ |
| Korean | ✓ |
| Persian a | |
| Polish | ✓ |
| Portugese | ✓ |
| Pushto | ✓ |
| Romanian | |
| Russian | ✓ |
| Spanish | ✓ |
| Swedish | |
| Thai | |
| Turkish | |
| Urdu | ✓ |
a. Western Farsi and Dari ↩