Simple methods for NER, including vanilla LSTM, LSTM + GloVe embedding, CRF.
- Please download and install the following:
- Python-crfsuites (https://python-crfsuite.readthedocs.io/en/latest/index.html)
- Pytorch (https://pytorch.org)
- Scikit-learn (https://scikit-learn.org/stable/)
- Pretrained GloVe (http://nlp.stanford.edu/data/glove.6B.zip) vectors, and put them under "data" folder
To reproduce the experiment, simply download the iPython notebooks in this repository and run all cells.
Note: for privacy issue, the data used in this base is restricted.