Repairing outlier behaviour in event logs

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

doi:10.1007/978-3-319-93931-5_9

Repairing outlier behaviour in event logs

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

Mathematics and Computer Science

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

15 Citations (Scopus)

Abstract

One of the main challenges in applying process mining on real event data, is the presence of noise and rare behaviour. Applying process mining algorithms directly on raw event data typically results in complex, incomprehensible, and, in some cases, even inaccurate analyses. As a result, correct and/or important behaviour may be concealed. In this paper, we propose an event data repair method, that tries to detect and repair outlier behaviour within the given event data. We propose a probabilistic method that is based on the occurrence frequency of activities in specific contexts. Our approach allows for removal of infrequent behaviour, which enables us to obtain a more global view of the process. The proposed method has been implemented in both the ProM- and the RapidProM framework. Using these implementations, we conduct a collection of experiments that show that we are able to detect and modify most types of outlier behaviour in the event data. Our evaluation clearly demonstrates that we are able to help to improve process mining discovery results by repairing event logs upfront.

Original language	English
Title of host publication	Business Information Systems - 21st International Conference, BIS 2018, Proceedings
Editors	W. Abramowicz, A. Paschke
Place of Publication	Cham
Publisher	Springer
Pages	115-131
Number of pages	17
ISBN (Electronic)	978-3-319-93931-5
ISBN (Print)	978-3-319-93930-8
DOIs	https://doi.org/10.1007/978-3-319-93931-5_9
Publication status	Published - 1 Jan 2018
Event	21st International Conference on Business Information Systems, (BIS 2018) - Berlin, Germany Duration: 18 Jul 2018 → 20 Jul 2018 https://link.springer.com/conference/bis

Publication series

Name	Lecture Notes in Business Information Processing
Volume	320
ISSN (Print)	1865-1348

Conference

Conference	21st International Conference on Business Information Systems, (BIS 2018)
Abbreviated title	BIS2018
Country/Territory	Germany
City	Berlin
Period	18/07/18 → 20/07/18
Internet address	https://link.springer.com/conference/bis

Keywords

Conditional probability
Data cleansing
Event log preprocessing
Log repair
Outlier detection
Process mining

Access to Document

10.1007/978-3-319-93931-5_9

Cite this

@inproceedings{5996973c956945f690e33672456955f3,

title = "Repairing outlier behaviour in event logs",

abstract = "One of the main challenges in applying process mining on real event data, is the presence of noise and rare behaviour. Applying process mining algorithms directly on raw event data typically results in complex, incomprehensible, and, in some cases, even inaccurate analyses. As a result, correct and/or important behaviour may be concealed. In this paper, we propose an event data repair method, that tries to detect and repair outlier behaviour within the given event data. We propose a probabilistic method that is based on the occurrence frequency of activities in specific contexts. Our approach allows for removal of infrequent behaviour, which enables us to obtain a more global view of the process. The proposed method has been implemented in both the ProM- and the RapidProM framework. Using these implementations, we conduct a collection of experiments that show that we are able to detect and modify most types of outlier behaviour in the event data. Our evaluation clearly demonstrates that we are able to help to improve process mining discovery results by repairing event logs upfront.",

keywords = "Conditional probability, Data cleansing, Event log preprocessing, Log repair, Outlier detection, Process mining",

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

year = "2018",

month = jan,

day = "1",

doi = "10.1007/978-3-319-93931-5_9",

language = "English",

isbn = "978-3-319-93930-8",

series = "Lecture Notes in Business Information Processing",

publisher = "Springer",

pages = "115--131",

editor = "W. Abramowicz and A. Paschke",

booktitle = "Business Information Systems - 21st International Conference, BIS 2018, Proceedings",

address = "Germany",

note = "21st International Conference on Business Information Systems, (BIS 2018), BIS2018 ; Conference date: 18-07-2018 Through 20-07-2018",

url = "https://link.springer.com/conference/bis",

}

Fani Sani, M, van Zelst, SJ & van der Aalst, WMP 2018, Repairing outlier behaviour in event logs. in W Abramowicz & A Paschke (eds), Business Information Systems - 21st International Conference, BIS 2018, Proceedings. Lecture Notes in Business Information Processing, vol. 320, Springer, Cham, pp. 115-131, 21st International Conference on Business Information Systems, (BIS 2018), Berlin, Germany, 18/07/18. https://doi.org/10.1007/978-3-319-93931-5_9

Repairing outlier behaviour in event logs. / Fani Sani, Mohammadreza; van Zelst, Sebastiaan J.; van der Aalst, Wil M.P.

Business Information Systems - 21st International Conference, BIS 2018, Proceedings. ed. / W. Abramowicz; A. Paschke. Cham : Springer, 2018. p. 115-131 (Lecture Notes in Business Information Processing; Vol. 320).

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

TY - GEN

T1 - Repairing outlier behaviour in event logs

AU - Fani Sani, Mohammadreza

AU - van Zelst, Sebastiaan J.

AU - van der Aalst, Wil M.P.

PY - 2018/1/1

Y1 - 2018/1/1

N2 - One of the main challenges in applying process mining on real event data, is the presence of noise and rare behaviour. Applying process mining algorithms directly on raw event data typically results in complex, incomprehensible, and, in some cases, even inaccurate analyses. As a result, correct and/or important behaviour may be concealed. In this paper, we propose an event data repair method, that tries to detect and repair outlier behaviour within the given event data. We propose a probabilistic method that is based on the occurrence frequency of activities in specific contexts. Our approach allows for removal of infrequent behaviour, which enables us to obtain a more global view of the process. The proposed method has been implemented in both the ProM- and the RapidProM framework. Using these implementations, we conduct a collection of experiments that show that we are able to detect and modify most types of outlier behaviour in the event data. Our evaluation clearly demonstrates that we are able to help to improve process mining discovery results by repairing event logs upfront.

AB - One of the main challenges in applying process mining on real event data, is the presence of noise and rare behaviour. Applying process mining algorithms directly on raw event data typically results in complex, incomprehensible, and, in some cases, even inaccurate analyses. As a result, correct and/or important behaviour may be concealed. In this paper, we propose an event data repair method, that tries to detect and repair outlier behaviour within the given event data. We propose a probabilistic method that is based on the occurrence frequency of activities in specific contexts. Our approach allows for removal of infrequent behaviour, which enables us to obtain a more global view of the process. The proposed method has been implemented in both the ProM- and the RapidProM framework. Using these implementations, we conduct a collection of experiments that show that we are able to detect and modify most types of outlier behaviour in the event data. Our evaluation clearly demonstrates that we are able to help to improve process mining discovery results by repairing event logs upfront.

KW - Conditional probability

KW - Data cleansing

KW - Event log preprocessing

KW - Log repair

KW - Outlier detection

KW - Process mining

UR - http://www.scopus.com/inward/record.url?scp=85050614184&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-93931-5_9

DO - 10.1007/978-3-319-93931-5_9

M3 - Conference contribution

AN - SCOPUS:85050614184

SN - 978-3-319-93930-8

T3 - Lecture Notes in Business Information Processing

SP - 115

EP - 131

BT - Business Information Systems - 21st International Conference, BIS 2018, Proceedings

A2 - Abramowicz, W.

A2 - Paschke, A.

PB - Springer

CY - Cham

T2 - 21st International Conference on Business Information Systems, (BIS 2018)

Y2 - 18 July 2018 through 20 July 2018

ER -

Repairing outlier behaviour in event logs

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this