Knowledge-Driven Learning for XAI in Fraud Detection

In this project, we develop an innovative approach to combating money laundering, based on advanced AI methods using hypergraph representations and process mining. This combination provides an integrated overview of the transactions involved, can handle the complex nature of the data, and remains interpretable for experts analysing the data, enabling them to make well-founded decisions.

  • PhD Student: Adia Lumadjeng
  • Supervisors: Erman Acar (IvI/ILLC), Ilker Birbil (ABS)
  • External Partner: ABN AMRO

Cases

Similar projects & activities

Bridging Tabular, Graph and Relational Foundation models

Mechanistic Interpretability for Tabular Transformers

ELSA lab AI for Health Equity

Advancing Machine Learning Methods for High-Frequency Financial Time-Series Data