Titlebar

Export bibliographic data
Literature by the same author
plus on the publication server
plus at Google Scholar

 

Machine Learning in Business Process Monitoring : A Comparison of Deep Learning and Classical Approaches Used for Outcome Prediction

DOI zum Zitieren der Version auf EPub Bayreuth: https://doi.org/10.15495/EPub_UBT_00005095
URN to cite this document: urn:nbn:de:bvb:703-epub-5095-9

Title data

Kratsch, Wolfgang ; Manderscheid, Jonas ; Röglinger, Maximilian ; Seyfried, Johannes:
Machine Learning in Business Process Monitoring : A Comparison of Deep Learning and Classical Approaches Used for Outcome Prediction.
In: Business & Information Systems Engineering. (2020) .
ISSN 1867-0202
DOI der Verlagsversion: https://doi.org/10.1007/s12599-020-00645-0

[img]
Format: PDF
Name: Kratsch2020_Article_MachineLearningInBusinessProce.pdf
Version: Published Version
Available under License Creative Commons BY 4.0: Attribution
Download (676kB)

Abstract

Predictive process monitoring aims at forecasting the behavior, performance, and outcomes of business processes at runtime. It helps identify problems before they occur and re-allocate resources before they are wasted. Although deep learning (DL) has yielded breakthroughs, most existing approaches build on classical machine learning (ML) techniques, particularly when it comes to outcome-oriented predictive process monitoring. This circumstance reflects a lack of understanding about which event log properties facilitate the use of DL techniques. To address this gap, the authors compared the performance of DL (i.e., simple feedforward Deep Neural Networks and Long Short Term Memory Networks) and ML tech-niques (i.e., Random Forests and Support Vector Machines) based on five publicly available event logs. It could be observed that DL generally outperforms classical ML techniques. Moreover, three specific propositions could be inferred from further observations: First, the outperformance of DL techniques is particularly strong for logs with a high variant-to-instance ratio (i.e., many non-standard cases). Sec-ond, DL techniques perform more stably in case of imbalanced target variables, especially for logs with a high event-to-activity ratio (i.e., many loops in the control flow). Third, logs with a high activity-to-instance payload ratio (i.e., input data is predominantly generated at runtime) call for the application of Long Short Term Memory Networks. Due to the purposive sampling of event logs and techniques, these findings also hold for logs outside this study.

Further data

Item Type: Article in a journal
Keywords: Predictive process monitoring; business process management; outcome prediction; deep learning; machine learning
DDC Subjects: 000 Computer Science, information, general works > 004 Computer science
300 Social sciences > 330 Economics
Institutions of the University: Faculties > Faculty of Law, Business and Economics > Department of Business Administration
Faculties > Faculty of Law, Business and Economics > Department of Business Administration > Professor Information Systems and Sustainable IT Management
Faculties > Faculty of Law, Business and Economics > Department of Business Administration > Chair Information Systems and Value-Based Business Process Management
Faculties > Faculty of Law, Business and Economics > Department of Business Administration > Chair Information Systems and Value-Based Business Process Management > Chair Information Systems and Value-Based Business Process Management - Univ.-Prof. Dr. Maximilian Röglinger
Research Institutions
Research Institutions > Affiliated Institutes
Research Institutions > Affiliated Institutes > Fraunhofer Project Group Business and Information Systems Engineering
Research Institutions > Affiliated Institutes > FIM Research Center Finance & Information Management
Faculties
Faculties > Faculty of Law, Business and Economics
Language: English
Originates at UBT: Yes
URN: urn:nbn:de:bvb:703-epub-5095-9
Date Deposited: 23 Sep 2020 08:21
Last Modified: 23 Sep 2020 08:21
URI: https://epub.uni-bayreuth.de/id/eprint/5095

Downloads

Downloads per month over past year