Improving process discovery results by filtering outliers using conditional behavioural probabilities

Mohammadreza Fani Sani; Sebastiaan J. van Zelst; Wil M.P. van der Aalst

doi:10.1007/978-3-319-74030-0_16

Improving process discovery results by filtering outliers using conditional behavioural probabilities

Mohammadreza Fani Sani, Sebastiaan J. van Zelst, Wil M.P. van der Aalst

Process Analytics

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

52 Citations (Scopus)

Abstract

Process discovery, one of the key challenges in process mining, aims at discovering process models from process execution data stored in event logs. Most discovery algorithms assume that all data in an event log conform to correct execution of the process, and hence, incorporate all behaviour in their resulting process model. However, in real event logs, noise and irrelevant infrequent behaviour are often present. Incorporating such behaviour results in complex, incomprehensible process models concealing the correct and/or relevant behaviour of the underlying process. In this paper, we propose a novel general purpose filtering method that exploits observed conditional probabilities between sequences of activities. The method has been implemented in both the ProM toolkit and the RapidProM framework. We evaluate our approach using real and synthetic event data. The results show that the proposed method accurately removes irrelevant behaviour and, indeed, improves process discovery results.

Original language	English
Title of host publication	Business Process Management Workshops - BPM 2017 International Workshops, Revised Papers
Place of Publication	Berlin
Publisher	Springer
Pages	216-229
Number of pages	14
ISBN (Print)	9783319740294
DOIs	https://doi.org/10.1007/978-3-319-74030-0_16
Publication status	Published - 2018
Event	15th International Conference on Business Process Management (BPM 2017) - Barcelona, Spain Duration: 10 Sept 2017 → 15 Sept 2017 Conference number: 15 https://bpm2017.cs.upc.edu/

Publication series

Name	Lecture Notes in Business Information Processing
Volume	308
ISSN (Print)	1865-1348

Conference

Conference	15th International Conference on Business Process Management (BPM 2017)
Abbreviated title	BPM 2017
Country/Territory	Spain
City	Barcelona
Period	10/09/17 → 15/09/17
Internet address	https://bpm2017.cs.upc.edu/

Keywords

Noise filtering
Outlier detection
Process discovery
Process mining
Noise filtering Outlier detection

Access to Document

10.1007/978-3-319-74030-0_16

Cite this

Sani, M. F., van Zelst, S. J., & van der Aalst, W. M. P. (2018). Improving process discovery results by filtering outliers using conditional behavioural probabilities. In Business Process Management Workshops - BPM 2017 International Workshops, Revised Papers (pp. 216-229). (Lecture Notes in Business Information Processing; Vol. 308). Springer. https://doi.org/10.1007/978-3-319-74030-0_16

@inproceedings{f623eec51a45418ca141799a21f31135,

title = "Improving process discovery results by filtering outliers using conditional behavioural probabilities",

abstract = "Process discovery, one of the key challenges in process mining, aims at discovering process models from process execution data stored in event logs. Most discovery algorithms assume that all data in an event log conform to correct execution of the process, and hence, incorporate all behaviour in their resulting process model. However, in real event logs, noise and irrelevant infrequent behaviour are often present. Incorporating such behaviour results in complex, incomprehensible process models concealing the correct and/or relevant behaviour of the underlying process. In this paper, we propose a novel general purpose filtering method that exploits observed conditional probabilities between sequences of activities. The method has been implemented in both the ProM toolkit and the RapidProM framework. We evaluate our approach using real and synthetic event data. The results show that the proposed method accurately removes irrelevant behaviour and, indeed, improves process discovery results.",

keywords = "Noise filtering, Outlier detection, Process discovery, Process mining, Noise filtering Outlier detection",

author = "Sani, {Mohammadreza Fani} and {van Zelst}, {Sebastiaan J.} and {van der Aalst}, {Wil M.P.}",

year = "2018",

doi = "10.1007/978-3-319-74030-0_16",

language = "English",

isbn = "9783319740294",

series = "Lecture Notes in Business Information Processing",

publisher = "Springer",

pages = "216--229",

booktitle = "Business Process Management Workshops - BPM 2017 International Workshops, Revised Papers",

address = "Germany",

note = "15th International Conference on Business Process Management (BPM 2017), BPM 2017 ; Conference date: 10-09-2017 Through 15-09-2017",

url = "https://bpm2017.cs.upc.edu/",

}

Sani, MF, van Zelst, SJ & van der Aalst, WMP 2018, Improving process discovery results by filtering outliers using conditional behavioural probabilities. in Business Process Management Workshops - BPM 2017 International Workshops, Revised Papers. Lecture Notes in Business Information Processing, vol. 308, Springer, Berlin, pp. 216-229, 15th International Conference on Business Process Management (BPM 2017), Barcelona, Spain, 10/09/17. https://doi.org/10.1007/978-3-319-74030-0_16

Improving process discovery results by filtering outliers using conditional behavioural probabilities. / Sani, Mohammadreza Fani; van Zelst, Sebastiaan J.; van der Aalst, Wil M.P.
Business Process Management Workshops - BPM 2017 International Workshops, Revised Papers. Berlin: Springer, 2018. p. 216-229 (Lecture Notes in Business Information Processing; Vol. 308).

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

TY - GEN

T1 - Improving process discovery results by filtering outliers using conditional behavioural probabilities

AU - Sani, Mohammadreza Fani

AU - van Zelst, Sebastiaan J.

AU - van der Aalst, Wil M.P.

N1 - Conference code: 15

PY - 2018

Y1 - 2018

N2 - Process discovery, one of the key challenges in process mining, aims at discovering process models from process execution data stored in event logs. Most discovery algorithms assume that all data in an event log conform to correct execution of the process, and hence, incorporate all behaviour in their resulting process model. However, in real event logs, noise and irrelevant infrequent behaviour are often present. Incorporating such behaviour results in complex, incomprehensible process models concealing the correct and/or relevant behaviour of the underlying process. In this paper, we propose a novel general purpose filtering method that exploits observed conditional probabilities between sequences of activities. The method has been implemented in both the ProM toolkit and the RapidProM framework. We evaluate our approach using real and synthetic event data. The results show that the proposed method accurately removes irrelevant behaviour and, indeed, improves process discovery results.

AB - Process discovery, one of the key challenges in process mining, aims at discovering process models from process execution data stored in event logs. Most discovery algorithms assume that all data in an event log conform to correct execution of the process, and hence, incorporate all behaviour in their resulting process model. However, in real event logs, noise and irrelevant infrequent behaviour are often present. Incorporating such behaviour results in complex, incomprehensible process models concealing the correct and/or relevant behaviour of the underlying process. In this paper, we propose a novel general purpose filtering method that exploits observed conditional probabilities between sequences of activities. The method has been implemented in both the ProM toolkit and the RapidProM framework. We evaluate our approach using real and synthetic event data. The results show that the proposed method accurately removes irrelevant behaviour and, indeed, improves process discovery results.

KW - Noise filtering

KW - Outlier detection

KW - Process discovery

KW - Process mining

KW - Noise filtering Outlier detection

UR - http://www.scopus.com/inward/record.url?scp=85041716900&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-74030-0_16

DO - 10.1007/978-3-319-74030-0_16

M3 - Conference contribution

AN - SCOPUS:85041716900

SN - 9783319740294

T3 - Lecture Notes in Business Information Processing

SP - 216

EP - 229

BT - Business Process Management Workshops - BPM 2017 International Workshops, Revised Papers

PB - Springer

CY - Berlin

T2 - 15th International Conference on Business Process Management (BPM 2017)

Y2 - 10 September 2017 through 15 September 2017

ER -

Improving process discovery results by filtering outliers using conditional behavioural probabilities

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this