Raising the Bar in Graph-level Anomaly Detection (GLAD)

This is the companion code for a PyTorch implementation of graph-level anomaly detection methods described in the paper Raising the Bar in Graph-level Anomaly Detection by Chen Qiu et al. The paper is published in IJCAI 2022 and can be found here https://arxiv.org/abs/2205.13845. The code allows the users to reproduce and extend the results reported in the study. Please cite the above paper when reporting, reproducing or extending the results.

Purpose of the project

This software is a research prototype, solely developed for and published as part of the publication cited above. It will neither be maintained nor monitored in any way.

Reproduce the Results

This repo contains the code of experiments with five methods (OCGTL,GTL,OCGIN,GTP,OCPool) on six graph datatsets.

Please run the command and replace $# with available options (see below):

python Launch_Exps.py --config-file $1 --dataset-name $2

config-file:

config_OCGTL.yml; config_OCGIN.yml; config_GTL.yml; config_GTP.yml; config_OCPool.yml

dataset-name:

dd; thyroid; nci1; aids; imdb; reddit;

How to Use

When using your own data, please put your data files under DATA.
Create a config file which contains your hyper-parameters under config_files.
Add your data loader to the loader/GraphDataClass.py.

Datasets

Graph Data are downloaded from TUDataset https://chrsmrrs.github.io/datasets/. Please put the data under DATA.

License

Raising the Bar in Graph-level Anomaly Detection (GLAD) is open-sourced under the AGPL-3.0 license. See the LICENSE file for details.