The final year project implements Variational Graph Autoencoder (VGAE) that reconstruct the connectivity of graph but also including node features. The reconstructed output is tested against 3 graph-based fraud classifier compared with original data.
- Create
MyDrive/data science project/in your google drive
- Download Dataset and save it to
./data/elliptic_txs_features/: https://www.kaggle.com/datasets/ellipticco/elliptic-data-set $ py dir_init.py$ pip install -r req.txt- run notebook