TU Darmstadt / ULB / TUbiblio

Unsupervised Anomaly Detection in Noisy Business Process Event Logs Using Denoising Autoencoders

Nolle, Timo ; Seeliger, Alexander ; Mühlhäuser, Max
eds.: Calders, Toon ; Ceci, Michelangelo ; Malerba, Donato (2016)
Unsupervised Anomaly Detection in Noisy Business Process Event Logs Using Denoising Autoencoders.
Bari, Italy
doi: 10.1007/978-3-319-46307-0_28
Conference or Workshop Item, Bibliographie

Abstract

Business processes are prone to subtle changes over time, as unwanted behavior slowly manifests in the execution flow. This problem is related to anomaly detection, as these subtle changes start of as anomalies at first, and thus it is important to detect them early. However, the necessary process documentation is often outdated, and thus not usable. Moreover, the only way of analyzing a process in execution is the use of event logs coming from process-aware information systems, but these event logs already contain anomalous behavior and other sorts of noise. Classic process anomaly detection algorithms require a dataset that is free of anomalies; thus, they are unable to process the noisy event logs. Within this paper we propose a system, relying on neural network technology, that is able to deal with the noise in the event log and learn a representation of the underlying model, and thus detect anomalous behavior based on this representation. We evaluate our approach on five different event logs, coming from process models with different complexities, and demonstrate that our approach yields remarkable results of 97.2 percent F1-score in detecting anomalous traces in the event log, and 95.6 percent accuracy in detecting the respective anomalous activities within the traces.

Item Type: Conference or Workshop Item
Erschienen: 2016
Editors: Calders, Toon ; Ceci, Michelangelo ; Malerba, Donato
Creators: Nolle, Timo ; Seeliger, Alexander ; Mühlhäuser, Max
Type of entry: Bibliographie
Title: Unsupervised Anomaly Detection in Noisy Business Process Event Logs Using Denoising Autoencoders
Language: German
Date: October 2016
Publisher: Calders, Toon Ceci, Michelangelo Malerba, Donato
Book Title: Discovery Science: 19th International Conference, DS 2016, Bari, Italy, Proceedings
Event Location: Bari, Italy
DOI: 10.1007/978-3-319-46307-0_28
Abstract:

Business processes are prone to subtle changes over time, as unwanted behavior slowly manifests in the execution flow. This problem is related to anomaly detection, as these subtle changes start of as anomalies at first, and thus it is important to detect them early. However, the necessary process documentation is often outdated, and thus not usable. Moreover, the only way of analyzing a process in execution is the use of event logs coming from process-aware information systems, but these event logs already contain anomalous behavior and other sorts of noise. Classic process anomaly detection algorithms require a dataset that is free of anomalies; thus, they are unable to process the noisy event logs. Within this paper we propose a system, relying on neural network technology, that is able to deal with the noise in the event log and learn a representation of the underlying model, and thus detect anomalous behavior based on this representation. We evaluate our approach on five different event logs, coming from process models with different complexities, and demonstrate that our approach yields remarkable results of 97.2 percent F1-score in detecting anomalous traces in the event log, and 95.6 percent accuracy in detecting the respective anomalous activities within the traces.

Identification Number: TUD-CS-2016-0168
Divisions: 20 Department of Computer Science
20 Department of Computer Science > Telecooperation
LOEWE
LOEWE > LOEWE-Schwerpunkte
LOEWE > LOEWE-Schwerpunkte > NiCER – Networked infrastructureless Cooperation for Emergency Response
Date Deposited: 31 Dec 2016 12:59
Last Modified: 14 Jun 2021 06:14
PPN:
Export:
Suche nach Titel in: TUfind oder in Google
Send an inquiry Send an inquiry

Options (only for editors)
Show editorial Details Show editorial Details