Repository navigation
Supported Systems Languages
| Platform | Status | Notes |
|---|---|---|
| Windows 10/11 (64-bit) | Fully supported | Standalone build available |
| macOS (Apple Silicon — M1/M2/M3/M4/M5) | Fully supported | Standalone build available |
| macOS (Intel) | Run from source | No standalone build currently provided; may be added if needed |
| Linux | Run from source | Not tested extensively; should work with the developer setup |
The NLP Suite supports 60+ languages through integration with multiple NLP packages:
| Package | Languages | Used for |
|---|---|---|
| Stanza | 60+ languages | Tokenization, POS tagging, NER, lemmatization, dependency parsing, sentiment analysis |
| spaCy | 20+ languages | Tokenization, POS tagging, NER, lemmatization |
| BERT | English + Multilingual | Sentiment analysis (English and multilingual models) |
Some dictionary-based algorithms are available only for English:
- WordNet (noun/verb aggregation)
- ANEW (sentiment/arousal/dominance)
- Hedonometer (sentiment)
- VADER (sentiment)
- NRC Emotion Wheel (8-emotion analysis)
- SentiWordNet (sentiment)
- Style analysis (concreteness, iconicity)
The NLP Suite language is configured in the Setup NLP package and corpus language dialog, accessible from the Setup menu at the bottom of most GUIs. The default is English with Stanza.
){kind=logo}