Graph Autoencoders for Business Process Anomaly Detection
siyu huo, Hagen Völzer, et al.
BPM 2021
The transactions of goods and services between enterprises are often driven by purchase orders (POs) and invoices. Every month thousands of invoices are billed to customers who settle them. Datasets across POs and invoices that are otherwise in silos need to be mapped for faster and more efficient operational analysis. In addition, the invoices may get into a dispute due to over exhaustion of allocated funds or may be billed to an expired PO. Managing the billing service is a huge undertaking along with increased costs. In this work, we develop an AI-driven purchase order management system to automate PO-invoice mapping and a risk analytics algorithm to decrease disputes as well as increase renewals.
siyu huo, Hagen Völzer, et al.
BPM 2021
Amadou Ba, Christopher Lohse, et al.
INFORMS 2022
Neil Thompson, Martin Fleming, et al.
IAAI 2024
Owen Cornec, Rahul Nair, et al.
NeurIPS 2021