Dialogue Discourse Parsing

Code for the AAAI-19 paper "A Deep Sequential Model for Discourse Parsing on Multi-Party Dialogues" (Shi and Huang, 2019)

Paper

this paper is now available on arXiv, and please kindly cite it:

@inproceedings{shi2019deep,
  title={A Deep Sequential Model for Discourse Parsing on Multi-Party Dialogues},
  author={Shi, Zhouxing and Huang, Minlie},
  booktitle={AAAI},
  year={2019}
}

Requirements

• Python 3

Dataset

We use the STAC corpus.

You may transform the original dataset into JSON files by:

python data_pre.py <input_dir> <output_json_file>

We also use 100-dimensional Glove word vectors.

Run

python main.py {--[option1]=[value1] --[option2]=[value2] ... }

Available options can be found at the top of main.py.

For example, to train the model with default settings:

python main.py --is_train