CasFlow

This repo provides a reference implementation of CasFlow as described in the paper:

CasFlow: Exploring Hierarchical Structures and Propagation Uncertainty for Cascade Prediction.
Fan Zhou, Xovee Xu, Kunpeng Zhang, Siyuan Liu and Goce Trajcevski.
Submitted for publication.

Basic Usage

Requirements

The code was tested with Python 3.7, tensorflow-gpu 2.0.0a and Cuda 10.0. Install the dependencies:

pip install -r requirements.txt

How to run the code

> cd casflow
> python gene_cascade.py
> python preprocessing.py
> python casflow.py

How to run the code via Google Colab

You can also run the code in Google Golab.

Options

You may change the model settings manually in config.py or directly into the codes.

Datasets

See some sample cascades in ./datasets/sample_data.txt.

The datasets we used in the paper can be obtained here:

• Twitter (Weng et al., Virality Prediction and Community Structure in Social Network, Scientific Report, 2013).
• Weibo (Cao et al., DeepHawkes: Bridging the Gap between Prediction and Understanding of Information Cascades., CIKM, 2017). You can also download Weibo dataset here in Google Drive.
• APS (Released by American Physical Society, obtained at Jan 17, 2019).

Cite

If you find CasFlow useful for your research, please consider citing us:

@inproceedings{zhou2020exploring,  
  author = {Zhou, Fan and Xu, Xovee and Zhang, Kunpeng and Liu, Siyuan and Trajcevski, Goce},  
  title = {Exploring Hierarchical Structures and Propagation Uncertainty for Cascade Prediction},
  booktitle = {Submitted for publication},
  year = {2020}
}