Portuguese Named Entity Recognition for Biomedical Information Extraction

This project provides models for the Named Entity Recognition task and algorithms to train, test, and evaluate the proposed models.

The data are in the Portuguese language and from the biomedical domain. We provided only one file as a data example, but we will release the full dataset soon.

For the full documentation, please follow this link: Docs.

Installation

First, check that you have Python 3.7+:

python3 --version

Then, you can install this package with the following instructions:

git clone https://github.com/pavalucas/portuguese-ner-biomedical.git
cd portuguese-ner-biomedical
python3 setup.py install

Tests

We developed tests using PyTest. All test files are located under the tests/ folder.

To run all tests:

pytest .

To output the test log to a file:

pytest . > testlog.txt

---

License

Distributed under the MIT License. See LICENSE.txt for more information.

Contact

Lucas Pavanelli - [email protected]

Project Link: portuguese-ner-biomedical