Using minimum description length for process mining

T.G.K. Calders; C.W. Günther; M. Pechenizkiy; A. Rozinat

doi:10.1145/1529282.1529606

Using minimum description length for process mining

T.G.K. Calders
, C.W. Günther
, M. Pechenizkiy
, A. Rozinat

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

45 Citations (Scopus)

Abstract

In the field of process mining, the goal is to automatically extract process models from event logs. Recently, many algorithms have been proposed for this task. For comparing these models, different quality measures have been proposed. Most of these measures, however, have several disadvantages; they are model-dependent, assume that the model that generated the log is known, or need negative examples of event sequences. In this paper we propose a new measure, based on the minimal description length principle, to evaluate the quality of process models that does not have these disadvantages. To illustrate the properties of the new measure we conduct experiments and discuss the trade-off between model complexity and compression.

Original language	English
Title of host publication	Proceedings of the 2009 ACM Symposium on Applied Computing (SAC 2009, Honolulu HI, USA, March 8-12, 2009)
Editors	S.Y. Shin, S. Ossowski
Place of Publication	New York NY
Publisher	Association for Computing Machinery, Inc.
Pages	1451-1455
ISBN (Print)	978-1-60558-166-8
DOIs	https://doi.org/10.1145/1529282.1529606
Publication status	Published - 2009
Event	24th ACM Symposium on Applied Computing (SAC 2009) - Hilton Hawaiian Village Beach Resort & Spa Waikiki Beach, Honolulu, United States Duration: 9 Mar 2009 → 12 Mar 2009 Conference number: 24

Conference

Conference	24th ACM Symposium on Applied Computing (SAC 2009)
Abbreviated title	SAC 2009
Country/Territory	United States
City	Honolulu
Period	9/03/09 → 12/03/09

Access to Document

10.1145/1529282.1529606

Cite this

Calders, T. G. K., Günther, C. W., Pechenizkiy, M., & Rozinat, A. (2009). Using minimum description length for process mining. In S. Y. Shin, & S. Ossowski (Eds.), Proceedings of the 2009 ACM Symposium on Applied Computing (SAC 2009, Honolulu HI, USA, March 8-12, 2009) (pp. 1451-1455). Association for Computing Machinery, Inc.. https://doi.org/10.1145/1529282.1529606

@inproceedings{63fdabe397c34196a246a47ab1b84b47,

title = "Using minimum description length for process mining",

abstract = "In the field of process mining, the goal is to automatically extract process models from event logs. Recently, many algorithms have been proposed for this task. For comparing these models, different quality measures have been proposed. Most of these measures, however, have several disadvantages; they are model-dependent, assume that the model that generated the log is known, or need negative examples of event sequences. In this paper we propose a new measure, based on the minimal description length principle, to evaluate the quality of process models that does not have these disadvantages. To illustrate the properties of the new measure we conduct experiments and discuss the trade-off between model complexity and compression.",

author = "T.G.K. Calders and C.W. G{\"u}nther and M. Pechenizkiy and A. Rozinat",

year = "2009",

doi = "10.1145/1529282.1529606",

language = "English",

isbn = "978-1-60558-166-8",

pages = "1451--1455",

editor = "S.Y. Shin and S. Ossowski",

booktitle = "Proceedings of the 2009 ACM Symposium on Applied Computing (SAC 2009, Honolulu HI, USA, March 8-12, 2009)",

publisher = "Association for Computing Machinery, Inc.",

address = "United States",

note = "24th ACM Symposium on Applied Computing (SAC 2009), SAC 2009 ; Conference date: 09-03-2009 Through 12-03-2009",

}

Calders, TGK, Günther, CW, Pechenizkiy, M & Rozinat, A 2009, Using minimum description length for process mining. in SY Shin & S Ossowski (eds), Proceedings of the 2009 ACM Symposium on Applied Computing (SAC 2009, Honolulu HI, USA, March 8-12, 2009). Association for Computing Machinery, Inc., New York NY, pp. 1451-1455, 24th ACM Symposium on Applied Computing (SAC 2009), Honolulu, Hawaii, United States, 9/03/09. https://doi.org/10.1145/1529282.1529606

Using minimum description length for process mining. / Calders, T.G.K.; Günther, C.W.; Pechenizkiy, M. et al.
Proceedings of the 2009 ACM Symposium on Applied Computing (SAC 2009, Honolulu HI, USA, March 8-12, 2009). ed. / S.Y. Shin; S. Ossowski. New York NY: Association for Computing Machinery, Inc., 2009. p. 1451-1455.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Using minimum description length for process mining

AU - Calders, T.G.K.

AU - Günther, C.W.

AU - Pechenizkiy, M.

AU - Rozinat, A.

N1 - Conference code: 24

PY - 2009

Y1 - 2009

N2 - In the field of process mining, the goal is to automatically extract process models from event logs. Recently, many algorithms have been proposed for this task. For comparing these models, different quality measures have been proposed. Most of these measures, however, have several disadvantages; they are model-dependent, assume that the model that generated the log is known, or need negative examples of event sequences. In this paper we propose a new measure, based on the minimal description length principle, to evaluate the quality of process models that does not have these disadvantages. To illustrate the properties of the new measure we conduct experiments and discuss the trade-off between model complexity and compression.

AB - In the field of process mining, the goal is to automatically extract process models from event logs. Recently, many algorithms have been proposed for this task. For comparing these models, different quality measures have been proposed. Most of these measures, however, have several disadvantages; they are model-dependent, assume that the model that generated the log is known, or need negative examples of event sequences. In this paper we propose a new measure, based on the minimal description length principle, to evaluate the quality of process models that does not have these disadvantages. To illustrate the properties of the new measure we conduct experiments and discuss the trade-off between model complexity and compression.

U2 - 10.1145/1529282.1529606

DO - 10.1145/1529282.1529606

M3 - Conference contribution

SN - 978-1-60558-166-8

SP - 1451

EP - 1455

BT - Proceedings of the 2009 ACM Symposium on Applied Computing (SAC 2009, Honolulu HI, USA, March 8-12, 2009)

A2 - Shin, S.Y.

A2 - Ossowski, S.

PB - Association for Computing Machinery, Inc.

CY - New York NY

T2 - 24th ACM Symposium on Applied Computing (SAC 2009)

Y2 - 9 March 2009 through 12 March 2009

ER -

Using minimum description length for process mining

Abstract

Conference

Access to Document

Fingerprint

Cite this