Mini-project
Banking & Fintech
Illicit transaction detection on a Bitcoin graph
Find illegal transactions in a Bitcoin transaction graph using the graph structure as well as the features.
The brief
Criminals move money through chains of wallets, which are hard to see one transaction at a time. Using a labelled Bitcoin transaction graph, compare a model that uses only transaction features with a graph neural network that also uses neighbours, and show which helps more. Good looks like: time-based splits so the model faces future data, precision and recall on the rare illicit class, and an explanation of the graph patterns found.
What you will build
- Load the transaction graph and labels
- Split by time
- Train a feature-only baseline
- Train a graph neural network
- Compare on the illicit class
- Explain the patterns
Free, open data to build it with
Sign up free to get the data and links
It takes a minute with your email. AiTeky Pass (₹999/month or ₹9,999/year) is free for your first 30 days.