Extended entity recognition

There can be cases when the default four-type NER does not suffice, one needs a more fine-grained entity type system. Attila Novak developed a corpus and an entity recognition system consisting more than 30 entity types. We provide easy integration with his tool.

Load

Loading the model can be achieved by adding the nerpp component.

import huspacy.integrations

nlp.add_pipe("nerpp")

Please note that using this component requires transformers, torch and spacy-alignments to be installed. Installing HuSpaCy with trf extras installs all these dependencies: pip install huspacy[trf]

Get entity annotations

The nerpp components stores entities as spans on the document under the "ents" key:

doc = nlp("A Ford Focus egy alsó-középkategóriás családi autó")
print(doc.spans["ents"])
print(doc.spans["ents"][0].label_) 

gives

[Ford Focus]
CAR

Citing

If you use this component, please cite:

@InProceedings{novak-novak:2022:LREC,
  author    = {Nov{\'{a}}k, Attila  and  Nov{\'{a}}k, Borb{\'{a}}la},
  title     = {NerKor+Cars-OntoNotes++},
  booktitle      = {Proceedings of the 13th Language Resources and Evaluation Conference (LREC 2022)},
  month          = {June},
  year           = {2022},
  address        = {Marseille, France},
  publisher      = {European Language Resources Association},
  pages     = {1907--1916},
    url       = {http://www.lrec-conf.org/proceedings/lrec2022/pdf/2022.lrec-1.203.pdf}
}