#TEC: A Time Evolving Contextual Graph Model for Speaker State Analysis in Political Debates

This codebase contains the python scripts for TEC, the model for the IJCAI 2021 paper TEC: A Time Evolving Contextual Graph Model for Speaker State Analysis in Political Debates

Environment & Installation Steps

Python 3.6, Pytorch, Pytorch-Geometric and networkx.

pip install -r requirements.txt

Dataset and Preprocessing

Download the dataset and follow preprocessing steps from here.

Run

Execute the following python command to train TEC:

python train_model.py

Cite

Consider citing our work if you use our codebase

@inproceedings{ijcai2021-0489,
  title     = {TEC: A Time Evolving Contextual Graph Model for Speaker State Analysis in Political Debates},
  author    = {Sawhney, Ramit and Agarwal, Shivam and Wadhwa, Arnav and Shah, Rajiv},
  booktitle = {Proceedings of the Thirtieth International Joint Conference on
               Artificial Intelligence, {IJCAI-21}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  editor    = {Zhi-Hua Zhou},
  pages     = {3552--3558},
  year      = {2021},
  month     = {8},
  note      = {Main Track}
  doi       = {10.24963/ijcai.2021/489},
  url       = {https://doi.org/10.24963/ijcai.2021/489},
}