/ner_brazilian_legal_texts

Named Entity Recognition in Brazilian Legal Texts using Conditional Random Fields and Word Embeddings

Primary LanguagePython

Named Entity Recognition in Brazilian Legal Texts using Conditional Random Fields and Word Embeddings

Running

  • Download LeNER-Br dataset using the script available in lener directory.

    • cd lener && sh download_dataset.sh

  • Run main.py, specifying the NER method and model:

    • python main.py --method CRF --model [FirstOrderCRF|HOCRFAD|HOSemiCRFAD]