Improving Declarative Process Mining with a Priori Noise Filtering

Publikation: Bidrag til bog/antologi/rapport › Konferencebidrag i proceedings › Forskning › fagfællebedømt

Standard

Improving Declarative Process Mining with a Priori Noise Filtering. / Christfort, Axel Kjeld Fjelrad; Debois, Søren; Slaats, Tijs.

Business Process Management Workshops - BPM 2022 International Workshops, Revised Selected Papers. red. / Cristina Cabanillas; Niels Frederik Garmann-Johnsen; Agnes Koschmider. Springer, 2023. s. 286-297 (Lecture Notes in Business Information Processing, Bind 460 LNBIP).

Publikation: Bidrag til bog/antologi/rapport › Konferencebidrag i proceedings › Forskning › fagfællebedømt

Harvard

Christfort, AKF, Debois, S & Slaats, T 2023, Improving Declarative Process Mining with a Priori Noise Filtering. i C Cabanillas, NF Garmann-Johnsen & A Koschmider (red), Business Process Management Workshops - BPM 2022 International Workshops, Revised Selected Papers. Springer, Lecture Notes in Business Information Processing, bind 460 LNBIP, s. 286-297, Workshops on AI4BPM, BP-Meet-IoT, BPI, BPM and RD, BPMS2, BPO, DEC2H, and NLP4BPM 2022, co-located with the 20th International Conference on Business Process Management, BPM 2022, Münster, Tyskland, 11/09/2022. https://doi.org/10.1007/978-3-031-25383-6_21

APA

Christfort, A. K. F., Debois, S., & Slaats, T. (2023). Improving Declarative Process Mining with a Priori Noise Filtering. I C. Cabanillas, N. F. Garmann-Johnsen, & A. Koschmider (red.), Business Process Management Workshops - BPM 2022 International Workshops, Revised Selected Papers (s. 286-297). Springer. Lecture Notes in Business Information Processing Bind 460 LNBIP https://doi.org/10.1007/978-3-031-25383-6_21

Vancouver

Christfort AKF, Debois S, Slaats T. Improving Declarative Process Mining with a Priori Noise Filtering. I Cabanillas C, Garmann-Johnsen NF, Koschmider A, red., Business Process Management Workshops - BPM 2022 International Workshops, Revised Selected Papers. Springer. 2023. s. 286-297. (Lecture Notes in Business Information Processing, Bind 460 LNBIP). https://doi.org/10.1007/978-3-031-25383-6_21

Author

Christfort, Axel Kjeld Fjelrad ; Debois, Søren ; Slaats, Tijs. / Improving Declarative Process Mining with a Priori Noise Filtering. Business Process Management Workshops - BPM 2022 International Workshops, Revised Selected Papers. red. / Cristina Cabanillas ; Niels Frederik Garmann-Johnsen ; Agnes Koschmider. Springer, 2023. s. 286-297 (Lecture Notes in Business Information Processing, Bind 460 LNBIP).

Bibtex

@inproceedings{f8091cd00257495cafd9f0f9d9d757a4,
title = "Improving Declarative Process Mining with a Priori Noise Filtering",
abstract = "In this paper, we report the results of an exploratory study into the efficacy of noise filtering in improving the accuracy of declarative process mining. We apply the double-granularity mixed-dependency filtering algorithm as introduced by [9], to the DisCoveR declarative miner [1], and parameter optimise it to only perform coarse-grained filtering. However, while noise filtering appears promising on the surface, one might worry that the outlier behaviour allowed by declarative models may be wrongly classified as noise and removed. To test the efficacy of noise filtering from both perspectives, we applied DisCoveR with noise filtering to two data sets: the process log collection used in the PDC2020 process discovery contest, emulating “real-world” scenarios; and a synthetic set of logs known to exhibit (non-noise) outlier behaviour. We find that on the contest data sets, noise filtering significantly increases accuracy (on average 23% points), obtaining exploratory evidence that noise filtering may improve declarative miner performance; on the synthetic logs we showcase examples where noise is filtered, while outlier behaviour remains untouched.",
keywords = "DCR Graphs, Declarative process models, Noise filtering, Process discovery",
author = "Christfort, {Axel Kjeld Fjelrad} and S{\o}ren Debois and Tijs Slaats",
note = "Publisher Copyright: {\textcopyright} 2023, Springer Nature Switzerland AG.; Workshops on AI4BPM, BP-Meet-IoT, BPI, BPM and RD, BPMS2, BPO, DEC2H, and NLP4BPM 2022, co-located with the 20th International Conference on Business Process Management, BPM 2022 ; Conference date: 11-09-2022 Through 16-09-2022",
year = "2023",
doi = "10.1007/978-3-031-25383-6_21",
language = "English",
isbn = "9783031253829",
series = "Lecture Notes in Business Information Processing",
publisher = "Springer",
pages = "286--297",
editor = "Cristina Cabanillas and Garmann-Johnsen, {Niels Frederik} and Agnes Koschmider",
booktitle = "Business Process Management Workshops - BPM 2022 International Workshops, Revised Selected Papers",
address = "Switzerland",

}

RIS

TY - GEN

T1 - Improving Declarative Process Mining with a Priori Noise Filtering

AU - Christfort, Axel Kjeld Fjelrad

AU - Debois, Søren

AU - Slaats, Tijs

N1 - Publisher Copyright: © 2023, Springer Nature Switzerland AG.

PY - 2023

Y1 - 2023

N2 - In this paper, we report the results of an exploratory study into the efficacy of noise filtering in improving the accuracy of declarative process mining. We apply the double-granularity mixed-dependency filtering algorithm as introduced by [9], to the DisCoveR declarative miner [1], and parameter optimise it to only perform coarse-grained filtering. However, while noise filtering appears promising on the surface, one might worry that the outlier behaviour allowed by declarative models may be wrongly classified as noise and removed. To test the efficacy of noise filtering from both perspectives, we applied DisCoveR with noise filtering to two data sets: the process log collection used in the PDC2020 process discovery contest, emulating “real-world” scenarios; and a synthetic set of logs known to exhibit (non-noise) outlier behaviour. We find that on the contest data sets, noise filtering significantly increases accuracy (on average 23% points), obtaining exploratory evidence that noise filtering may improve declarative miner performance; on the synthetic logs we showcase examples where noise is filtered, while outlier behaviour remains untouched.

AB - In this paper, we report the results of an exploratory study into the efficacy of noise filtering in improving the accuracy of declarative process mining. We apply the double-granularity mixed-dependency filtering algorithm as introduced by [9], to the DisCoveR declarative miner [1], and parameter optimise it to only perform coarse-grained filtering. However, while noise filtering appears promising on the surface, one might worry that the outlier behaviour allowed by declarative models may be wrongly classified as noise and removed. To test the efficacy of noise filtering from both perspectives, we applied DisCoveR with noise filtering to two data sets: the process log collection used in the PDC2020 process discovery contest, emulating “real-world” scenarios; and a synthetic set of logs known to exhibit (non-noise) outlier behaviour. We find that on the contest data sets, noise filtering significantly increases accuracy (on average 23% points), obtaining exploratory evidence that noise filtering may improve declarative miner performance; on the synthetic logs we showcase examples where noise is filtered, while outlier behaviour remains untouched.

KW - DCR Graphs

KW - Declarative process models

KW - Noise filtering

KW - Process discovery

UR - http://www.scopus.com/inward/record.url?scp=85151064083&partnerID=8YFLogxK

U2 - 10.1007/978-3-031-25383-6_21

DO - 10.1007/978-3-031-25383-6_21

M3 - Article in proceedings

AN - SCOPUS:85151064083

SN - 9783031253829

T3 - Lecture Notes in Business Information Processing

SP - 286

EP - 297

BT - Business Process Management Workshops - BPM 2022 International Workshops, Revised Selected Papers

A2 - Cabanillas, Cristina

A2 - Garmann-Johnsen, Niels Frederik

A2 - Koschmider, Agnes

PB - Springer

T2 - Workshops on AI4BPM, BP-Meet-IoT, BPI, BPM and RD, BPMS2, BPO, DEC2H, and NLP4BPM 2022, co-located with the 20th International Conference on Business Process Management, BPM 2022

Y2 - 11 September 2022 through 16 September 2022

ER -

ID: 343224343