This is the source code for paper ''Text-enhanced Multi-Granularity Temporal Graph Learning for Event Prediction'' appeared in the proceedings of ICDM 2022 (regular paper).

Xiaoxue Han, Yue Ning

• ICEWS event data is available online.
• ICEWS news data has not been released publicly.

The code has been successfully tested in the following environment. (For older versions, you may need to modify the code)

• Python 3.8.3
• PyTorch 1.10.0
• dgl 0.8.1
• Sklearn 0.23.1
• numpy 1.18.5

• THA (Event and news data in Bangkok, Thailand from 2010 to 2016) OneDrive

Please run the following commands for training and testing. We take the dataset THA as the example.

python train.py --data 'THA' -lt 1 -pw 1 -hw 7 --n_runs 1 

Please cite our paper if you find this code useful for your research:

X. Han and Y. Ning, "Text-enhanced Multi-Granularity Temporal Graph Learning for Event Prediction," 2022 IEEE International Conference on Data Mining (ICDM), Orlando, FL, USA, 2022, pp. 171-180, doi: 10.1109/ICDM54844.2022.00027.

BibTeX

@INPROCEEDINGS{10027692,
  author={Han, Xiaoxue and Ning, Yue},
  booktitle={2022 IEEE International Conference on Data Mining (ICDM)}, 
  title={Text-enhanced Multi-Granularity Temporal Graph Learning for Event Prediction}, 
  year={2022},
  volume={},
  number={},
  pages={171-180},
  doi={10.1109/ICDM54844.2022.00027}}