Site for testing a coreference resolution system on a user-friendly web interface.

Code was written in Python3.5.5, and you must use a Python3.5.x version to load the Tensorflow model.

It's advised to use a virtualenv.

Polyglot depends on numpy and libicu-dev.

On Linux use:

sudo apt-get install python-numpy libicu-dev

On macOS use:

brew install icu4c

CFLAGS="-I/usr/local/opt/icu4c/include -std=c++11" LDFLAGS=-L/usr/local/opt/icu4c/lib pip install pyicu

pip install -r requirements.txt

Start the backend server:

python application.py

Then navigate to 127.0.0.1:5000

This repository contains code used for the following publication (link):

@INPROCEEDINGS{8554705,
    author={A. F. Cruz and G. Rocha and H. L. Cardoso},
    booktitle={2018 Fifth International Conference on Social Networks Analysis, Management and Security (SNAMS)},
    title={Exploring Spanish Corpora for Portuguese Coreference Resolution},
    year={2018},
    pages={290-295},
    keywords={Task analysis;Training;Natural language processing;Measurement;Social network services;Security;Feature extraction},
    doi={10.1109/SNAMS.2018.8554705},
    month={Oct},
}