Jointly Multiple Events Extraction (JMEE)

This is the code of the Jointly Multiple Events Extraction (JMEE) in our EMNLP 2018 paper.

Updated Answers

1. We upload the data split files qi_filelist using in preprocessing with stanford corenlp.
2. We provide an example calling the runner for training.

Requirement

• python 3
• pytorch == 0.4.0
• torchtext == 0.2.3
• tensorboardX
• seqeval

To install the requirements, run pip -r requirements.txt.

How to run the code?

After preprocessing the ACE 2005 dataset and put it under ace-05-splits, the main entrance is in enet/run/ee/runner.py. We cannot include the data in this release due to licence issues.

But we offer a piece of data sample in ace-05-splits/sample.json, the format should be followed.

THE CODE IS A BASIC PRELIMINARY VERSION AND IS LIKE "AS IS", WITHOUT WARRANTY OF ANY KIND.

Cite

Please cite our EMNLP 2018 paper:

@inproceedings{DBLP:conf/emnlp/LiuLH18,
  author    = {Xiao Liu and
               Zhunchen Luo and
               Heyan Huang},
  title     = {Jointly Multiple Events Extraction via Attention-based Graph Information
               Aggregation},
  booktitle = {Proceedings of the 2018 Conference on Empirical Methods in Natural
               Language Processing, Brussels, Belgium, October 31 - November 4, 2018},
  pages     = {1247--1256},
  year      = {2018},
  crossref  = {DBLP:conf/emnlp/2018},
  url       = {https://aclanthology.info/papers/D18-1156/d18-1156},
  timestamp = {Sat, 27 Oct 2018 20:04:50 +0200},
  biburl    = {https://dblp.org/rec/bib/conf/emnlp/LiuLH18},
  bibsource = {dblp computer science bibliography, https://dblp.org}
}