search

uvacw / inca

24 stars 6 forks source link

Add spacy support #334

Open damian0604 opened 6 years ago

damian0604 commented 6 years ago

use spacy for POS-tagging, NER, etc. http://spacy.io

Available for Dutch, should be a good replacement for earlier processors that had more complicated depencies

damian0604 commented 6 years ago

and NER-tag everything

In [1]: import spacy

In [2]: nlp = spacy.load('nl')

In [3]: doc = nlp('De CEO van Siemens vindt het allemaal niks')

In [4]: doc.ents
Out[4]: (Siemens,)
damian0604 commented 6 years ago

PR #335 added NER. POS still to be done.