GitHub - FORMAS/DptOIE

DPToie is an Open Information Extraction method for Brazilian Portuguese language. It extracts triples based on the Universal Dependencies.

To run

Colab
Models

How to use the single Java

To run the DptOIE.jar

Insert the file DptOIE.jar and pt-models in your directory

 java -jar DptOIE.jar -sentencesIN **sentences_file_path**

To use the module that handles subordinate clause

To use the module that handles coordinated conjunctions

To use the module that handles appositive

To apply transitivity

DptOIE is can receive CONLL-U format to annotate the sentences with the dependency parser.

To run DptOIE from a dependency tree in ConLL-U format

java -jar DptOIE.jar -sentencesIN 'sentences_file_path' -dependencyTreeIN 'dependency_Tree_conllu_format'

Contributing

Please use the GitHub issue tracker

Authors

How to cite

If you find this repo helpful, please consider citing:

Leandro Oliveira, Daniela Barreiro Claro, Marlo Souza: DptOIE: a Portuguese open information extraction based on dependency analysis. Artif. Intell. Rev. 56(7): 7015-7046 (2023).

@article{DBLP:journals/air/OliveiraCS23,
  author       = {Leandro Oliveira and
                  Daniela Barreiro Claro and
                  Marlo Souza},
  title        = {DptOIE: a Portuguese open information extraction based on dependency
                  analysis},
  journal      = {Artif. Intell. Rev.},
  volume       = {56},
  number       = {7},
  pages        = {7015--7046},
  year         = {2023},
  url          = {https://doi.org/10.1007/s10462-022-10349-4},
  doi          = {10.1007/s10462-022-10349-4},
  timestamp    = {Thu, 15 Jun 2023 21:57:36 +0200},
  biburl       = {https://dblp.org/rec/journals/air/OliveiraCS23.bib},
  bibsource    = {dblp computer science bibliography, https://dblp.org}
}