©Leuphana

Research Colloquium on Business Informatics

02. Dec

As part of the Research Colloquium on Business Informatics and Data Science, Felix Krieger (Doktorand Leuphana Universität Lüneburg - IIS / Data Scientist in der Forschungs- und Entwicklungsabteilung von EY Deutschland im Bereich Wirtschaftsprüfung) will speak on 2 Dezember 2021, 16:15, about "Information Extraction from Invoices: Automating the Test of Details in Auditing".

Auditing is among the professions which have been predicted to be most affected by computerization, as it involves many repetitive and structured tasks. A prime example for such tasks is the ‘test of details’, which is used frequently in audits to collect evidence that transactions have been recorded correctly from invoices. However, the simplicity of the task is opposed by the complexity of invoice layouts and the virtually endless possibilities to design them. A growing body of research addresses this complexity through layout-aware machine learning approaches, allowing the extraction of structured information from layout-rich documents. A successful automation of the test of details based on these approaches could vastly improve the efficiency of audit field work, thereby also allowing auditors to focus on higher value-added tasks.

The research covered in this talk is concerned with the design of a machine learning backend for a ‘test of details’ solution, including a benchmark study on the effectiveness of different machine learning approaches. The design choices are further contextualized using a previously conducted study on technology adoption in the audit domain.

Access data

https://leuphana.zoom.us/j/99663552105?pwd=dG9KZUtJU0hUMkE4RHNEVDhiUlNIZz09

Meeting-ID: 99663552105
Password: FoKoWI