Enhancing process mining results using domain knowledge

P.M. Dixit, J.C.A.M. Buijs, W.M.P. van der Aalst, B.F.A. Hompes, J. Buurman

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

6 Citations (Scopus)
126 Downloads (Pure)

Abstract

Process discovery algorithms typically aim at discovering process models from event logs. Most discovery algorithms discover the model based on an event log, without allowing the domain expert to influence the discovery approach in any way. However, the user may have certain domain expertise which should be exploited to create a better process model. In this paper, we address this issue of incorporating domain knowledge to improve the discovered process model. We firstly present a modification algorithm to modify a discovered process model. Furthermore, we present a verification algorithm to verify the presence of user specified constraints in the model. The outcome of our approach is a Pareto front of process models based on the constraints specified by the domain expert and common quality dimensions of process mining.

Original languageEnglish
Title of host publicationInternational Symposium on Data-driven Process Discovery and Analysis 2015, Vienna, Austria.
EditorsP. Caravolo, S. Rinderle-Ma
PublisherCEUR-WS.org
Pages79-94
Number of pages16
Publication statusPublished - 2015
Event5th International Symposium on Data-Driven Process Discovery and Analysis (SIMPDA 2015 - Vienna, Austria
Duration: 9 Dec 201511 Dec 2015
Conference number: 5

Publication series

NameCEUR Workshop Proceedings
PublisherCEUR
Volume1527
ISSN (Print)1613-0073

Conference

Conference5th International Symposium on Data-Driven Process Discovery and Analysis (SIMPDA 2015
Abbreviated titleSIMPDA 2015
Country/TerritoryAustria
CityVienna
Period9/12/1511/12/15

Keywords

  • Algorithm Post Processing
  • Declare Templates
  • Domain Knowledge
  • User Guided Process Discovery

Fingerprint

Dive into the research topics of 'Enhancing process mining results using domain knowledge'. Together they form a unique fingerprint.

Cite this