Decomposed process mining with DivideAndConquer

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

4 Citations (Scopus)
63 Downloads (Pure)

Abstract

Many known process mining techniques scale badly in the number of activities in an event log. Examples of such techniques include the ILP Miner and the standard replay, which also uses ILP techniques. To alleviate the problems these techniques face, we can decompose a large problem (with many activities) into a number of small problems (with few activities). Expectation is, that the run times of such a decomposed setting will be faster than the run time of the original setting. This paper presents the DivideAndConquer tool, which allows the user to decompose a large problem into small problems, to run the desired discovery or replay technique on each of these decomposed problems, and to merge the results into a single result, which can then be shown to the user.
Original languageEnglish
Title of host publicationBPM Demo Sessions 2014 (co-located with BPM 2014, Eindhoven, The Netherlands, September 20, 2014)
EditorsL. Limonad, B. Weber
PublisherCEUR-WS.org
Pages86-90
Publication statusPublished - 2014
EventBPM Demo Sessions 2014 (BPMD 2014), September 10, 2014, Eindhoven, The Netherlands - Eindhoven, Netherlands
Duration: 10 Sep 201410 Sep 2014

Publication series

NameCEUR Workshop Proceedings
Volume1295
ISSN (Print)1613-0073

Other

OtherBPM Demo Sessions 2014 (BPMD 2014), September 10, 2014, Eindhoven, The Netherlands
Abbreviated titleBPMD 2014
CountryNetherlands
CityEindhoven
Period10/09/1410/09/14
OtherCo-located with the 12th International Conference on Business Process Management (BPM 2014)

Fingerprint

Dive into the research topics of 'Decomposed process mining with DivideAndConquer'. Together they form a unique fingerprint.

Cite this