Domain-specific Event Abstraction

Finn Klessascheck, Tom Lichtenstein, Martin Meier, Simon Remy, Jan Philipp Sachs, Luise Pufahl, Riccardo Miotto, Erwin Böttinger, Mathias Weske

Research output: Contribution to journalConference articlepeer-review

3 Scopus citations

Abstract

Process mining aims at deriving process knowledge from event logs, which contain data recorded during process executions. Typically, event logs need to be generated from process execution data, stored in different kinds of information systems. In complex domains like healthcare, data is available only at different levels of granularity. Event abstraction techniques allow the transformation of events to a common level of granularity, which enables effective process mining. Existing event abstraction techniques do not sufficiently take into account domain knowledge and, as a result, fail to deliver suitable event logs in complex application domains. This paper presents an event abstraction method based on domain ontologies. We show that the method introduced generates semantically meaningful high-level events, suitable for process mining; it is evaluated on real-world patient treatment data of a large U.S. health system.

Original languageEnglish
Pages (from-to)117-126
Number of pages10
JournalBusiness Information Systems
Volume1
DOIs
StatePublished - 2021
Externally publishedYes
Event24th International Conference on Business Information Systems, BIS 2021 - Hanover, Germany
Duration: 15 Jun 202117 Jun 2021

Keywords

  • Domain knowledge
  • Event abstraction
  • Healthcare
  • Process mining

Fingerprint

Dive into the research topics of 'Domain-specific Event Abstraction'. Together they form a unique fingerprint.

Cite this