hogunpark / tsgnet

TSGNet (Temporal-Static-Graph-Net)

TSGNet

This is a Pytorch implementation of TSGNet (Temporal-Static-Graph-Net), as described in our paper:

Hogun Park, Jennifer Neville, Exploiting Interaction Links for Node Classification with Deep Graph Neural Networks (IJCAI 2019)

Usage

Example Usage $main.py --dataname "imdb"

Full Command List The full list of command-line options is available with $main.py --help

Requirements

We tested the codes in the following environment, but it may work in their previous versions.

• Python 3.6
• Pytorch 1.7.0
• Python Geometric 1.6.3

Cite

If you find TSGNet useful in your research, we ask that you cite the following paper::

@inproceedings{parkneville2019,
  title     = {Exploiting Interaction Links for Node Classification with Deep Graph Neural Networks},
  author    = {Park, Hogun and Neville, Jennifer},
  booktitle = {Proceedings of the Twenty-Eighth International Joint Conference on
               Artificial Intelligence, {IJCAI-19}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  pages     = {3223--3230},
  year      = {2019},
  month     = {7},
  doi       = {10.24963/ijcai.2019/447},
  url       = {https://doi.org/10.24963/ijcai.2019/447},
}

Misc

Datasets and Implementation of layers are following interfaces of Pytorch geometric.

tsgnet's Issues

Request for Additional Dataset: Facebook

Hello！

I hope you're doing well. I recently came across your paper and code on GitHub. I'm impressed with your work and would like to request an additional dataset.

I'm currently working on replicating and extending your research using the provided dataset. However, I believe that including a Facebook dataset for comparative experiments would greatly enhance the validity and scope of my findings.

I assure you that my intention is purely for academic research, and I have no commercial or non-academic use in mind. I understand the importance of data privacy and will handle the dataset with the utmost care.

Could you kindly provide me with an additional dataset related to Facebook that aligns with the objectives of your paper? This would greatly contribute to the advancement of research in this field and help me validate the effectiveness of your methods.

Thank you for considering my request. Your support in providing the dataset would be highly appreciated.