RST Discourse Parser

This is the code for the paper : Nan Yu, Meishan Zhang, Guohong Fu. Transition-based Neural RST Parsing with Implicit Syntax Features. COLING'18, 2018.

How to compile this project in Linux.

• Step 0: Open terminal, and change directory to project directory.
  Use this command cd /your/project/path/NNDisParser.
  
• Step 1: Create a new directory in NNDisParser.
  Use this command mkdir build.
  
• Step 2: Change your directory. Use this command cd build.
  
• Step 3: Use this command to build project.
  cmake .. -DEIGEN3_DIR=${path}eigen -DN3L_DIR=${path}N3LDG -DMKL=TRUE.
  
• Step 4: Now you can compile this project by command make.
  
• Step 5: If you want to run this project. And add this argument.
  -train /your/training/corpus -dev /your/development/corpus -test /your/test/corpus -option /your/option/file -l
  

Data

RST Tree Bank.
https://catalog.ldc.upenn.edu/LDC2002T07

External Resource

Pretrained word embeddings.
https://nlp.stanford.edu/projects/glove

Biaffine parser hidden.
https://drive.google.com/drive/folders/1has4GJ98RUGKRDB0slHDseWF9dJ4JrNg?usp=sharing

NOTE

Make sure you have eigen ,N3LDG, cmake.
Eigen:http://eigen.tuxfamily.org/index.php?title=Main_Page
cmake:https://cmake.org