Avoiding over-fitting in ILP-based process discovery

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

20 Citations (Scopus)
1 Downloads (Pure)

Abstract

The aim of process discovery is to discover a process model based on business process execution data, recorded in an event log. One of several existing process discovery techniques is the ILP-based process discovery algorithm. The algorithm is able to unravel complex process structures and provides formal guarantees w.r.t. the model discovered, e.g., the algorithm guarantees that a discovered model describes all behavior present in the event log. Unfortunately the algorithm is unable to cope with exceptional behavior present in event logs. As a result, the application of ILP-based process discovery techniques in everyday process discovery practice is limited. This paper addresses this problem by proposing a filtering technique tailored towards ILP-based process discovery. The technique helps to produce process models that are less over-fitting w.r.t. the event log, more understandable, and more adequate in capturing the dominant behavior present in the event log. The technique is implemented in the ProM framework. Keywords: Process mining Process discovery Integer linear programming Filtering
Original languageEnglish
Title of host publicationBusiness Process Management
Subtitle of host publication13th International Conference, BPM 2015, Innsbruck, Austria, August 31 - September 3, 2015, Proceedings
EditorsH.R. Motahari-Nezhad, J. Recker, M. Weidlich
Place of PublicationDordrecht
PublisherSpringer
Pages163-171
ISBN (Electronic)978-3-319-23063-4
ISBN (Print)978-3-319-23062-7
DOIs
Publication statusPublished - 2015

Publication series

NameLecture Notes in Computer Science
Volume9253
ISSN (Print)0302-9743

Fingerprint

Inductive logic programming (ILP)
Linear programming
Industry

Cite this

Zelst, van, S. J., Dongen, van, B. F., & van der Aalst, W. M. P. (2015). Avoiding over-fitting in ILP-based process discovery. In H. R. Motahari-Nezhad, J. Recker, & M. Weidlich (Eds.), Business Process Management : 13th International Conference, BPM 2015, Innsbruck, Austria, August 31 - September 3, 2015, Proceedings (pp. 163-171). (Lecture Notes in Computer Science; Vol. 9253). Dordrecht: Springer. https://doi.org/10.1007/978-3-319-23063-4_10
Zelst, van, S.J. ; Dongen, van, B.F. ; van der Aalst, W.M.P. / Avoiding over-fitting in ILP-based process discovery. Business Process Management : 13th International Conference, BPM 2015, Innsbruck, Austria, August 31 - September 3, 2015, Proceedings. editor / H.R. Motahari-Nezhad ; J. Recker ; M. Weidlich. Dordrecht : Springer, 2015. pp. 163-171 (Lecture Notes in Computer Science).
@inproceedings{65bf728d16474015b7c3757c84b2ed33,
title = "Avoiding over-fitting in ILP-based process discovery",
abstract = "The aim of process discovery is to discover a process model based on business process execution data, recorded in an event log. One of several existing process discovery techniques is the ILP-based process discovery algorithm. The algorithm is able to unravel complex process structures and provides formal guarantees w.r.t. the model discovered, e.g., the algorithm guarantees that a discovered model describes all behavior present in the event log. Unfortunately the algorithm is unable to cope with exceptional behavior present in event logs. As a result, the application of ILP-based process discovery techniques in everyday process discovery practice is limited. This paper addresses this problem by proposing a filtering technique tailored towards ILP-based process discovery. The technique helps to produce process models that are less over-fitting w.r.t. the event log, more understandable, and more adequate in capturing the dominant behavior present in the event log. The technique is implemented in the ProM framework. Keywords: Process mining Process discovery Integer linear programming Filtering",
author = "{Zelst, van}, S.J. and {Dongen, van}, B.F. and {van der Aalst}, W.M.P.",
year = "2015",
doi = "10.1007/978-3-319-23063-4_10",
language = "English",
isbn = "978-3-319-23062-7",
series = "Lecture Notes in Computer Science",
publisher = "Springer",
pages = "163--171",
editor = "H.R. Motahari-Nezhad and J. Recker and M. Weidlich",
booktitle = "Business Process Management",
address = "Germany",

}

Zelst, van, SJ, Dongen, van, BF & van der Aalst, WMP 2015, Avoiding over-fitting in ILP-based process discovery. in HR Motahari-Nezhad, J Recker & M Weidlich (eds), Business Process Management : 13th International Conference, BPM 2015, Innsbruck, Austria, August 31 - September 3, 2015, Proceedings. Lecture Notes in Computer Science, vol. 9253, Springer, Dordrecht, pp. 163-171. https://doi.org/10.1007/978-3-319-23063-4_10

Avoiding over-fitting in ILP-based process discovery. / Zelst, van, S.J.; Dongen, van, B.F.; van der Aalst, W.M.P.

Business Process Management : 13th International Conference, BPM 2015, Innsbruck, Austria, August 31 - September 3, 2015, Proceedings. ed. / H.R. Motahari-Nezhad; J. Recker; M. Weidlich. Dordrecht : Springer, 2015. p. 163-171 (Lecture Notes in Computer Science; Vol. 9253).

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

TY - GEN

T1 - Avoiding over-fitting in ILP-based process discovery

AU - Zelst, van, S.J.

AU - Dongen, van, B.F.

AU - van der Aalst, W.M.P.

PY - 2015

Y1 - 2015

N2 - The aim of process discovery is to discover a process model based on business process execution data, recorded in an event log. One of several existing process discovery techniques is the ILP-based process discovery algorithm. The algorithm is able to unravel complex process structures and provides formal guarantees w.r.t. the model discovered, e.g., the algorithm guarantees that a discovered model describes all behavior present in the event log. Unfortunately the algorithm is unable to cope with exceptional behavior present in event logs. As a result, the application of ILP-based process discovery techniques in everyday process discovery practice is limited. This paper addresses this problem by proposing a filtering technique tailored towards ILP-based process discovery. The technique helps to produce process models that are less over-fitting w.r.t. the event log, more understandable, and more adequate in capturing the dominant behavior present in the event log. The technique is implemented in the ProM framework. Keywords: Process mining Process discovery Integer linear programming Filtering

AB - The aim of process discovery is to discover a process model based on business process execution data, recorded in an event log. One of several existing process discovery techniques is the ILP-based process discovery algorithm. The algorithm is able to unravel complex process structures and provides formal guarantees w.r.t. the model discovered, e.g., the algorithm guarantees that a discovered model describes all behavior present in the event log. Unfortunately the algorithm is unable to cope with exceptional behavior present in event logs. As a result, the application of ILP-based process discovery techniques in everyday process discovery practice is limited. This paper addresses this problem by proposing a filtering technique tailored towards ILP-based process discovery. The technique helps to produce process models that are less over-fitting w.r.t. the event log, more understandable, and more adequate in capturing the dominant behavior present in the event log. The technique is implemented in the ProM framework. Keywords: Process mining Process discovery Integer linear programming Filtering

U2 - 10.1007/978-3-319-23063-4_10

DO - 10.1007/978-3-319-23063-4_10

M3 - Conference contribution

SN - 978-3-319-23062-7

T3 - Lecture Notes in Computer Science

SP - 163

EP - 171

BT - Business Process Management

A2 - Motahari-Nezhad, H.R.

A2 - Recker, J.

A2 - Weidlich, M.

PB - Springer

CY - Dordrecht

ER -

Zelst, van SJ, Dongen, van BF, van der Aalst WMP. Avoiding over-fitting in ILP-based process discovery. In Motahari-Nezhad HR, Recker J, Weidlich M, editors, Business Process Management : 13th International Conference, BPM 2015, Innsbruck, Austria, August 31 - September 3, 2015, Proceedings. Dordrecht: Springer. 2015. p. 163-171. (Lecture Notes in Computer Science). https://doi.org/10.1007/978-3-319-23063-4_10