Skip to content
ACoLi Shallow Discourse Parser
Branch: master
Clone or download
Nikoschenk Update README.md
- changed sample call to label implicit relations.
Latest commit 79aba3c Oct 10, 2016
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
implicit_nn_training Update README.md Oct 10, 2016
README.md

README.md

The Frankfurt Shallow Discourse Parser

Developed at the Applied Computational Linguistics Lab (ACoLi), Goethe University Frankfurt am Main, Germany.

This repository hosts the shallow discourse parser described in: Do We Really Need All Those Rich Linguistic Features? A Neural Network-Based Approach to Implicit Sense Labeling

Niko Schenk, Christian Chiarcos, Samuel Rönnqvist, Kathrin Donandt, Evgeny A. Stepanov and Giuseppe Riccardi. "Do We Really Need All Those Rich Linguistic Features? A Neural Network-Based Approach to Implicit Sense Labeling". In Proceedings of the Twentieth Conference on Computational Natural Language Learning - Shared Task, CoNLL 2016. 2016.

@inproceedings{schenk-EtAl:2016:CoNLL-STSDP,
  author    = {Niko Schenk, Christian Chiarcos, Samuel Rönnqvist, Kathrin Donandt, Evgeny A. Stepanov,  Giuseppe Riccardi},
  title     = {{Do We Really Need All Those Rich Linguistic Features? A Neural Network-Based Approach to Implicit Sense Labeling}},
  booktitle = {Proceedings of the Twentieth Conference on Computational Natural Language Learning - Shared Task, CoNLL 2016},
  month     = {August},
  year      = {2016},
  address   = {Berlin, Germany},
  publisher = {Association for Computational Linguistics}
}

Software Requirements

The parser runs on a Linux environment using Python 2.7.

Please install the following prerequisites in the given order:

$ pip install Cython 
$ pip install --upgrade gensim
$ pip install theanets
$ pip install nose-parameterized
$ pip install -U scikit-learn

If you encounter issues with the numpy version, you can install all the software on a virtual environment (http://docs.python-guide.org/en/latest/dev/virtualenvs/):

$ pip install virtualenv
$ cd shallow-discourse-parser
$ virtualenv venv
$ source venv/bin/activate
$ pip install Cython
$ pip install --upgrade gensim
$ pip install theanets
$ pip install nose-parameterized
$ pip install -U scikit-learn

Data Requirements

Please copy the following data into the data/ directory:

For English:

For Chinese:

You can’t perform that action at this time.