The official implementation of [AHEAD: A Triple Attention Based Heterogeneous Graph Anomaly Detection Approach](https://arxiv.org/abs/2208.08200) -CIAC 2023 Best Paper Candidate
Hi, I am very interested in your work and would like to adopt my own dataset to use AHEAD model. But it looks like CoAID.mat and IMDB.mat are processed versions of original datasets and it is not clear what X1, X2 and edge_index keys in the datasets stands for.
Are X1 and X2 attribute matrices for 2 types of nodes? How were they generated?
What is edge_index?