Case ID Revealed HERE: Hybrid Elusive Case Repair Method for Transformer-Driven Business Process Event Log Enhancement

Felix Zetzsche (), Robert Andrews, Arthur H. M. Hofstede, Maximilian Röglinger, Sebastian Johannes Schmid and Moe Thandar Wynn
Additional contact information
Felix Zetzsche: FIM Research Center for Information Management, University of Bayreuth, Branch Business and Information Systems Engineering of the Fraunhofer FIT
Robert Andrews: Queensland University of Technology
Arthur H. M. Hofstede: Queensland University of Technology
Maximilian Röglinger: FIM Research Center for Information Management, University of Bayreuth, Branch Business and Information Systems Engineering of the Fraunhofer FIT
Sebastian Johannes Schmid: FIM Research Center for Information Management, University of Bayreuth, Branch Business and Information Systems Engineering of the Fraunhofer FIT
Moe Thandar Wynn: Queensland University of Technology

Business & Information Systems Engineering: The International Journal of WIRTSCHAFTSINFORMATIK, 2025, vol. 67, issue 3, No 2, 337 pages

Abstract: Abstract Process mining is a data-driven technique that leverages event logs to analyze, visualize, and improve business processes. However, data quality is often low in real-world settings due to various event log imperfections, which, in turn, degrade the accuracy and reliability of process mining insights. One notable example is the elusive case imperfection pattern, describing the absence of case identifiers responsible for linking events to a specific process instance. Elusive cases are particularly problematic, as process mining techniques rely heavily on the accurate mapping of events to instances to provide meaningful and actionable insights into business processes. To address this issue, the study follows the Design Science Research paradigm to iteratively develop a method for repairing the elusive case imperfection pattern in event logs. The proposed Hybrid Elusive Case Repair Method (HERE) combines a traditional, rule-based approach with generative artificial intelligence, specifically the Transformer architecture. By integrating domain knowledge, HERE constitutes a comprehensive human-in-the-loop approach, enhancing its ability to accurately repair elusive cases in event logs. The method is evaluated by instantiating it as a software prototype, applying it to repair three publicly accessible event logs, and seeking expert feedback in a total of 21 interviews conducted at different points during the design and development phase. The results demonstrate that HERE makes significant progress in addressing the elusive case imperfection pattern, particularly when provided with sufficient data volume, laying the groundwork for resolving further data quality issues in process mining.

Keywords: Process mining; Event log quality; Event log repair; Generative artificial intelligence; Transformer; Business process management (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://link.springer.com/10.1007/s12599-025-00935-5 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:binfse:v:67:y:2025:i:3:d:10.1007_s12599-025-00935-5

Ordering information: This journal article can be ordered from
http://www.springer.com/economics/journal/12599

DOI: 10.1007/s12599-025-00935-5

Access Statistics for this article

Business & Information Systems Engineering: The International Journal of WIRTSCHAFTSINFORMATIK is currently edited by Martin Bichler

More articles in Business & Information Systems Engineering: The International Journal of WIRTSCHAFTSINFORMATIK from Springer, Gesellschaft für Informatik e.V. (GI)
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().