RAGCL

Source code for RAGCL in paper:

Propagation Tree is not Deep: Adaptive Graph Contrastive Learning Approach for Rumor Detection

Run

You need to compute node centrality with centrality.py to process your own data before running the code.

The code can be run in the following ways:

nohup python main\(sup\).py --gpu 0 &

Dataset is available at https://pan.baidu.com/s/1Kl5IQjU3a_pdt90YmNNVsQ?pwd=qqul

Dependencies

• pytorch == 1.12.0

• torch-geometric == 2.1.0

• gensim == 4.0.1

Citation

If this work is helpful, please kindly cite as:

@inproceedings{cui2024propagation,
  title={Propagation Tree Is Not Deep: Adaptive Graph Contrastive Learning Approach for Rumor Detection},
  author={Cui, Chaoqun and Jia, Caiyan},
  booktitle={Proceedings of the AAAI Conference on Artificial Intelligence},
  volume={38},
  number={1},
  pages={73--81},
  year={2024}
}