SL-GAD

A PyTorch implementation of "Generative and Contrastive Self-Supervised Learning for Graph Anomaly Detection", IEEE Transactions on Knowledge and Data Engineering (TKDE).

Dependencies

• python==3.6.1
• dgl==0.4.1
• matplotlib==3.3.4
• networkx==2.5
• numpy==1.19.2
• pyparsing==2.4.7
• scikit-learn==0.24.1
• scipy==1.5.2
• sklearn==0.24.1
• torch==1.8.1
• tqdm==4.59.0

To install all dependencies:

pip install -r requirements.txt

Usage

To train and evaluate on BlogCatalog:

python run.py --device cuda:0 --expid 1 --dataset BlogCatalog --runs 5 --auc_test_rounds 256 --alpha 1.0 --beta 0.6

To train and evaluate on Flickr:

python run.py --device cuda:0 --expid 2 --dataset Flickr --runs 5 --auc_test_rounds 256 --alpha 1.0 --beta 0.6

To train and evaluate on Cora:

python run.py --device cuda:0 --expid 3 --dataset cora --runs 5 --auc_test_rounds 256 --alpha 1.0 --beta 0.6

To train and evaluate on CiteSeer:

python run.py --device cuda:0 --expid 4 --dataset citeseer --runs 5 --auc_test_rounds 256 --alpha 1.0 --beta 0.4

To train and evaluate on PubMed:

python run.py --device cuda:0 --expid 5 --dataset pubmed --runs 5 --auc_test_rounds 256 --alpha 1.0 --beta 0.4

To train and evaluate on ACM:

python run.py --device cuda:0 --expid 6 --dataset ACM --runs 5 --auc_test_rounds 256 --alpha 1.0 --beta 0.2

Citation

If you use our code in your research, please cite the following article:

@article{zheng2021generative,
  title={Generative and Contrastive Self-Supervised Learning for Graph Anomaly Detection},
  author={Zheng, Yu and Jin, Ming and Liu, Yixin and Chi, Lianhua and Phan, Khoa T and Chen, Yi-Ping Phoebe},
  journal={IEEE Transactions on Knowledge and Data Engineering (TKDE)},
  year={2021}
}