This is the source code for the paper "GAME-ON: Graph Attention Network based Multimodal Fusion for Fake News Detection."
-
creating-graph-data
contains the code for creating node embeddings for textual and visual graphs. -
unimodal-textual
,unimodal-visual
, andmultimodal-concatenation
folders contain code for ablation study. -
game-on
contains code for the proposed framework. -
ModelSizeBaselines.ipynb
is the code for roughly estimating the trainable paramters in the baselines.
-
MediaEval 2015 (Twitter Dataset): Link to image-verificatiion corpus github
-
Weibo Dataset: Link to EANN paper for the dataset
-
To run any of the available models, please move to the appropriate directory
-
run
$ python main.py