Skip to content

JiaWu-Repository/DeepSLRD

 
 

Repository files navigation

DeepSLRD

The implementation of our IJCNN 2019 paper "Deep Structure Learning for Rumor Detection on Twitter" DeepSLRD.

Requirements

python 3.6.6
numpy==1.17.2
networkx==2.2
scipy==1.3.1

How to use

Dataset

unzip dataset.zip
The dataset.zip includes nflod, resource, twitter15 and twitter16 folders. This dataset collected by Ma et al. (2018), and the raw datasets can be downloaded from here:
Jing Ma, Wei Gao, Kam-Fai Wong. Rumor Detection on Twitter with Tree-structured Recursive Neural Networks. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics, ACL 2018.
The ind_twitter15.graph, ind_twitter15.features, ind_twitter15.poster, ind_twitter16.graph, ind_twitter16.features, and ind_twitter16.poster files are the propocessed data of user behavious graph on datasets twitter15 and twitter16, respectively.

Training & Testing

python Main_BU_RvNN_GCN.py #training and testing the BU-Hybrid model
python Main_TD_RvNN_GCN.py #training and testing the TD-Hybrid model

Citation

If you find the code is useful for your research, please cite this paper:

@inproceedings{huang2019deep,
author = {Huang, Qi and Zhou, Chuan and Wu, Jia and Wang, Mingwen and Wang, Bin},
year = {2019},
month = {07},
pages = {1-8},
title = {Deep Structure Learning for Rumor Detection on Twitter},
doi = {10.1109/IJCNN.2019.8852468}
}

About

Deep Structure Learning for Rumor Detection on Twitter (IJCNN 2019)

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%