Skip to content

amiroft/RECON

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

11 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

RECON

This is the code for the paper 'RECON: Relation Extraction using Knowledge Graph Context in a Graph Neural Network'.

Please note, we recieved multiple queries regarding why we have not used BERT as context aggregator instead of GNN. We tried it and performance was quite low (in the range of 0.76 F1Score on Wikidata dataset). Hence we decided to ignore BERT.

Citation

If you use our work kindly consider citing

@inproceedings{bastos2020recon,
  title={RECON: Relation Extraction using Knowledge Graph Context in a Graph Neural Network},
  author={Bastos, Anson and Nadgeri, Abhishek and Singh, Kuldeep and Mulang, Isaiah Onando and Shekarpour, Saeedeh and Hoffart, Johannes and Kaul, Manohar},
  booktitle={Proceedings of The Web Conference (WWW) (long papers)},
  pages={:N/A},
  year={2021}
}

Dataset

The datasets can be downloaded from the below link

Wikidata

NYT

Setup

Requires python3, torch>=1.4.0

Running

Train

The parameters are in the config file. Please ensure all the file paths are appropriate and run

python train.py

Test

Ensure the right model is being loaded and run

python test.py

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%