mboillet's picture
Update README.md
c3e218a verified
metadata
library_name: Spacy
license: mit
tags:
  - Spacy
  - Named entity recognition
metrics:
  - P
  - R
  - F1
language:
  - la
  - de
  - cs
version:
  - Spacy v2

Spacy - HOME-NACR Multilingual

This model detects Person and Location entities in Latin, Czech and German.

Model description

The model has been trained using the Spacy v2 library on the HOME-NACR document annotations. The model is compatible with version 2.3.2 of Spacy and incompatible with versions 3.x.x.

Evaluation results

The model achieves the following results on HOME-NACR:

tag predicted matched Precision Recall F1 Support
PERS 28,276 28,006 0.990 0.997 0.994 28,087
LOC 27,541 27,165 0.986 0.987 0.987 27,528
All 55,817 55,171 0.988 0.992 0.990 55,615

How to use?

Please refer to the Spacy library to use this model.

Cite us!

@inproceedings{spacy2022,
    author = {Monroc, Claire Bizon and Miret, Blanche and Bonhomme, Marie-Laurence and Kermorvant, Christopher},
    title = {{A Comprehensive Study Of Open-Source Libraries For Named Entity Recognition On Handwritten Historical Documents}},
    year = {2022},
    isbn = {978-3-031-06554-5},
    url = {https://doi.org/10.1007/978-3-031-06555-2_29},
    doi = {10.1007/978-3-031-06555-2_29},
    booktitle = {Document Analysis Systems: 15th IAPR International Workshop, DAS 2022, La Rochelle, France, May 22–25, 2022, Proceedings},
    pages = {429–444},
    numpages = {16},
}