Code for AAAI 2025 paper --- Public Opinion Field Effect and Hawkes Process Join Hands for Information Popularity Prediction.
POFHP:.
│ get_data.py
│ pytorchtools.py
│ requirements.txt
│
├─data
│ ├─android
│ │
│ ├─christianity
│ │
│ ├─douban
│ │
│ └─twitter
│
└─src
│ main.py
│ models.py
│ util.py
│
└─ckpt- get_data.py: This file is used to process the data.
- pytorchtools.py: This file is used to define the early_stopping mechanism.
- requirements.txt: Dependencies file.
- data/：Dataset folder.
- src/main.py: Main file.
- src/models.py: POFHP implementation.
- src/util.py: Defining various toolkits.
Please install the following packages:
gensim==3.8.3
joblib==1.3.2
matplotlib==3.7.5
networkx==3.1
numpy==1.24.4
pandas==2.0.3
scikit-learn==1.3.2
scipy==1.10.1
torch==2.1.2+cu121
torch-cluster==1.6.3+pt21cu121
torch-geometric==2.5.2
torch-scatter==2.1.2+pt21cu121
torch-sparse==0.6.18+pt21cu121
torch-spline-conv==1.2.2+pt21cu121
tqdm==4.66.2
transformers==4.39.0
You can also simply run:
pip install -r requirements.txt
Since the Android and Twitter datasets are large, we compressed them. Therefore, the two datasets should be decompressed before use:
cd data/
unzip douban.zip
unzip twitter.zipThen:
cd src/
python main.py --data_name christianity
python main.py --data_name android
python main.py --data_name twitter
python main.py --data_name douban@inproceedings{li2025public,
title={Public Opinion Field Effect and Hawkes Process Join Hands for Information Popularity Prediction},
author={Li, Junliang and Yang, Yajun and Zhang, Yujia and Hu, Qinghua and Zhao, Alan and Gao, Hong},
booktitle={Proceedings of the AAAI Conference on Artificial Intelligence},
volume={39},
number={11},
pages={12076--12083},
year={2025}
}