This project implements the paper "Heterophily Learning and Global-local Dependencies Enhanced Multi-view Representation Learning for Graph Anomaly Detection" published in journal Knowledge-Based Systems(https://doi.org/10.1016/j.knosys.2025.114039).

This project is tested on cuda 11.6 with several dependencies listed below:

pytorch=1.11.0
torch-geometric=2.0.4

Public datasets weibo and toloker used for graph anomaly detection are available for evaluation.

python benchamrk.py --datasets 0/1