financial_transaction_scoring

Using graph embeddings and Tensorflow to predict AML fraud

Right now, I have a Colab that has a pretty basic model using an embedding of the transaction graph. This can run on a decent laptop. As I look through the simulated data, I might enhance the model; I haven't made a fraud-detection model before.

Run in Google Colab

Data

I am using some data from the IBM AMLSim repo, which has several datasets pre-generated and some nice papers:

https://github.com/IBM/AMLSim

As I familarized myself with this data, I found their wiki helpful. They have published two papers related to their repo. I recommend looking at this one:

Scalable Graph Learning for Anti-Money Laundering: A First Look, Mark Weber, et. al. 2018.

They cite other work that could be interesting too:

1. Vec2Struc: A Method Towards Explainable Structural-Based Node Embeddings
2. Anti-Money Laundering in Bitcoin: Experimenting with Graph Convolutional Networks for Financial Forensics

Predicted Variables

There are two targets of interest from the data on Dropbox:

1. IS_FRAUD: a binary label
2. ALERT_ID: a categorical label that's more descriptive