This is a PyTorch implementation of CMINet for the task of predicting social influence, as described in our paper: CMINet: a Graph Learning Framework for Content-aware Multi-channel Influence Diffusion
The code is tested on PyTorch 1.13, pytorch-geometric 2.1.0.
Please download them from the following links. Digg dataset Twitter dataset Flickr dataset
python3 preprocess.py --dataset <dataset-name>
python3 train.py --dataset <dataset> --argfile <file_path>
python3 train.py --dataset <dataset> --task <diffusion/activation> --argfile <file_path>