Scienceie17

NTNU at SemEval 2017 Task 10: Extracting Keyphrases and Relations from Scientific Publications

https://scienceie.github.io/

Setup

1. Install Anaconda Python distibution from https://www.continuum.io/downloads

2. Create virtual Python environment

   $ conda create -n scienceie17 python=3 scikit-learn

3. Activate environment

   $ source activate scienceie17

4. Clone git repos

   $ git clone https://github.com/OC-ScienceIE/scienceie17.git

5. Extend Python search path

   $ export PYTHONPATH=<.../scienceie17>/bin:<.../scienceie17>/lib/python:$PYTHON

   where <.../scienceie17> is the full path to the local repository.

6. Install additional dependencies

   $ pip install sklearn-crfsuite $ pip install -U spacy $ python -m spacy.en.download all

7. Preprocess data (run Spacy NLP, derive IOB tags)

   $ bin/preproc.py

For subsequent use, step 3 & 5 can also be carried out by running

$ source setup_env.sh