Generating Formal Process Models and Decisions from Examination Rules in Natural Text with ChatGPT

Andreas Speck, Melanie Windrich, Jan Hesse, Melina Sentz, David Kuhlen, Thomas Stuht, Elke Pulvermüller

2025

Abstract

Many business and administrative systems are modeled with business process models in a notation like Business Process Model and Notation (BPMN) supported by tools like Camunda (BPMN, 2025). When such systems are to be build in most cases the requirements and rules are recorded in plain text. This raises the desire using AI tools (artificial intelligence) for generating business process models from the text. The question is to which extend AI techniques may support the development of formal process models. We apply ChatGPT to analyze judicial regulations written natural text (examination regulation) and request transforming the text to process models and decision diagrams as an XML exchange file which may be displayed in the Camunda Modeler.

Paper Citation

in Harvard Style

Speck A., Windrich M., Hesse J., Sentz M., Kuhlen D., Stuht T. and Pulvermüller E. (2025). Generating Formal Process Models and Decisions from Examination Rules in Natural Text with ChatGPT. In Proceedings of the 20th International Conference on Evaluation of Novel Approaches to Software Engineering - Volume 1: ENASE; ISBN 978-989-758-742-9, SciTePress, pages 848-857. DOI: 10.5220/0013481600003928

in Bibtex Style

@conference{enase25,
author={Andreas Speck and Melanie Windrich and Jan Hesse and Melina Sentz and David Kuhlen and Thomas Stuht and Elke Pulvermüller},
title={Generating Formal Process Models and Decisions from Examination Rules in Natural Text with ChatGPT},
booktitle={Proceedings of the 20th International Conference on Evaluation of Novel Approaches to Software Engineering - Volume 1: ENASE},
year={2025},
pages={848-857},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013481600003928},
isbn={978-989-758-742-9},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 20th International Conference on Evaluation of Novel Approaches to Software Engineering - Volume 1: ENASE
TI - Generating Formal Process Models and Decisions from Examination Rules in Natural Text with ChatGPT
SN - 978-989-758-742-9
AU - Speck A.
AU - Windrich M.
AU - Hesse J.
AU - Sentz M.
AU - Kuhlen D.
AU - Stuht T.
AU - Pulvermüller E.
PY - 2025
SP - 848
EP - 857
DO - 10.5220/0013481600003928
PB - SciTePress