Source code for SIGIR 2021 paper Rumor Detection on Social Media with Event Augmentations
python 3.7
pytorch 1.8.1
pytorch_geometric 1.7.0
create "Twitter15graph" folder and "Twitter16graph" folder in the data folder
python ./Process/getTwittergraph.py Twitter15 # pre-process the Twitter15 dataset
python ./Process/getTwittergraph.py Twitter16 # pre-process the Twitter16 dataset
python ./Model/train.py Twitter15 100 # Run RDEA for 100 iterations on Twitter15 dataset
python ./Model/train.py Twitter16 100 # Run RDEA for 100 iterations on Twitter16 dataset
We use Twitter15 and Twitter16 dataset for the experiment.
To learn more about the dataset, please refer to RvNN for more details.
If you have any question, please contact zhenyu dot h at outlook dot com
If you find this code useful, please cite our paper.