KOEX

BMBF Project: Collaborative Machine Learning for the Detection of Fraud and Risks in ERP Systems (KOEX)

The widespread use of IT systems and in particular ERP systems for managing business processes has opened up a variety of new points of attack for fraudulent or erroneous behavior. While the focus of current security solutions is usually on external attacks, internal misuse, error, incorrect operation and fraud are often given secondary consideration. The Association of Certified Fraud Examiners estimates the damage caused to companies by fraud at around 5% of their annual turnover.

Current ERP systems, above all the SAP SE system, manage large amounts of data that provide information about the activities performed by employees as well as logging the entire flow of goods and finances of the company. Previous procedures usually use predefined rules to detect fraud in this data. However, new fraud cases are not detected by these approaches.

The goal of the KOEX project is to automatically detect known and unknown fraud cases in the data using machine learning methods. The identified fraud patterns are to be abstracted and made usable for other companies through federated learning techniques. In this way, no confidential details and personal data will be released. The insights gained from individual fraud cases can thus be used without drawing any conclusions about the source. The diverse competences of the partners involved are to be used to create an initial demonstration system within the scope of this project, which will be further developed and incorporated into the existing SIVIS suite after the end of the project.

Project homepage:
BMBF website of the KOEX project

Partners:
SIVIS GmbH (project management)
Prenode GmbH
Hochschule Karlsruhe

Status: Ongoing project 01/2022 - 06/2024

Contact:
Prof. Schmidt-Thieme
Johannes Burchert