This repository was is linked to work that was used in the paper - Detecting fake news spreaders in social networks using inductive representation learning. This paper was published in the IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM) 2020.
Please note that the code is still in a maintenance state.
If you find this code or paper useful in your research, please consider citing:
Please cite this paper:
@inproceedings{rath2020detecting,
title={Detecting fake news spreaders in social networks using inductive representation learning},
author={Rath, Bhavtosh and Salecha, Aadesh and Srivastava, Jaideep},
booktitle={2020 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM)},
pages={182--189},
year={2020},
organization={IEEE}
}
This project contains IGraph and NetworkX implementations of the following repo: https://github.com/BhavtoshRath/l1_community_ops
It also contains the latest implementations for TSM calculation based on: https://github.com/BhavtoshRath/TSM_python
The following functionality is not working as of 7.27.19:
a) second_level stats for IGraph implementation
b) spreader_role stats for both NetworkX and IGraph implementaion
c) Igraph implementation for bel_ranks is working (but needs to be validated against ground truth)