Exhaustive Model Identification on Process Mining

Takeharu Mitsuda, Hiroyuki Nakagawa, Hiroyuki Nakagawa, Haruhiko Kaiya, Hironori Takeuchi, Sinpei Ogata, Tatsuhiro Tsuchiya

2025

Abstract

HeuristicsMiner is a process mining technique, which can construct a process model representing dependency relations of each activity from event logs. HeuristicsMiner is notable for its ability to output a process model that removes noise from the input data by allowing the user to set multiple parameters. However, it is difficult for users to understand the characteristics of each parameter and to identify parameter values that enable them to obtain ideal process models. In this study, we propose a method for identifying all possible process models that can be generated from an input event log in HeuristicsMiner. We extract the conditions under which the dependencies in the input logs are represented in the output model, and then create a process model transition table based on these conditions to identify these models. We applied this method to several large logs and mined process models using the combinations of parameter values obtained, and confirmed that process models were efficiently obtained without excesses or deficiencies.

Paper Citation

in Harvard Style

Mitsuda T., Nakagawa H., Kaiya H., Takeuchi H., Ogata S. and Tsuchiya T. (2025). Exhaustive Model Identification on Process Mining. 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 449-456. DOI: 10.5220/0013270400003928

in Bibtex Style

@conference{enase25,
author={Takeharu Mitsuda and Hiroyuki Nakagawa and Haruhiko Kaiya and Hironori Takeuchi and Sinpei Ogata and Tatsuhiro Tsuchiya},
title={Exhaustive Model Identification on Process Mining},
booktitle={Proceedings of the 20th International Conference on Evaluation of Novel Approaches to Software Engineering - Volume 1: ENASE},
year={2025},
pages={449-456},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013270400003928},
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 - Exhaustive Model Identification on Process Mining
SN - 978-989-758-742-9
AU - Mitsuda T.
AU - Nakagawa H.
AU - Kaiya H.
AU - Takeuchi H.
AU - Ogata S.
AU - Tsuchiya T.
PY - 2025
SP - 449
EP - 456
DO - 10.5220/0013270400003928
PB - SciTePress