Show simple item record

dc.contributor.authorNoering, Fabian Kai Dietrich
dc.contributor.authorSchroeder, Yannik
dc.contributor.authorJonas, Konstantin
dc.contributor.authorKlawonn, Frank
dc.date.accessioned2021-07-27T12:32:41Z
dc.date.available2021-07-27T12:32:41Z
dc.date.issued2021-01-01
dc.identifier.citation(2021)Integrated Computer-Aided Engineering; 28(3): 237-256. DOI: 10.3233/ICA-210650.en_US
dc.identifier.issn10692509
dc.identifier.doi10.3233/ica-210650
dc.identifier.urihttp://hdl.handle.net/10033/622962
dc.description.abstractIn technical systems the analysis of similar situations is a promising technique to gain information about the system's state, its health or wearing. Very often, situations cannot be defined but need to be discovered as recurrent patterns within time series data of the system under consideration. This paper addresses the assessment of different approaches to discover frequent variable-length patterns in time series. Because of the success of artificial neural networks (NN) in various research fields, a special issue of this work is the applicability of NNs to the problem of pattern discovery in time series. Therefore we applied and adapted a Convolutional Autoencoder and compared it to classical nonlearning approaches based on Dynamic Time Warping, based on time series discretization as well as based on the Matrix Profile. These nonlearning approaches have also been adapted, to fulfill our requirements like the discovery of potentially time scaled patterns from noisy time series. We showed the performance (quality, computing time, effort of parametrization) of those approaches in an extensive test with synthetic data sets. Additionally the transferability to other data sets is tested by using real life vehicle data. We demonstrated the ability of Convolutional Autoencoders to discover patterns in an unsupervised way. Furthermore the tests showed, that the Autoencoder is able to discover patterns with a similar quality like classical nonlearning approaches. © 2021 - IOS Press. All rights reserved.en_US
dc.language.isoenen_US
dc.publisherIOS Pressen_US
dc.rightsAttribution 4.0 International*
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/*
dc.subjectautoencoderen_US
dc.subjectmotif discoveryen_US
dc.subjectpattern discoveryen_US
dc.subjectTime series data miningen_US
dc.subjectunsuperviseden_US
dc.titlePattern discovery in time series using autoencoder in comparison to nonlearning approachesen_US
dc.typeArticleen_US
dc.identifier.eissn18758835
dc.contributor.departmentHZI,Helmholtz-Zentrum für Infektionsforschung GmbH, Inhoffenstr. 7,38124 Braunschweig, Germany.en_US
dc.identifier.journalIntegrated Computer-Aided Engineeringen_US
dc.identifier.eid2-s2.0-85109132894
dc.identifier.scopusidSCOPUS_ID:85109132894
dc.source.volume28
dc.source.issue3
dc.source.beginpage237
dc.source.endpage256
refterms.dateFOA2021-07-27T12:32:42Z
dc.source.journaltitleIntegrated Computer-Aided Engineering


Files in this item

Thumbnail
Name:
Noering et al.pdf
Size:
15.17Mb
Format:
PDF
Description:
accepted manuscript

This item appears in the following Collection(s)

Show simple item record

Attribution 4.0 International
Except where otherwise noted, this item's license is described as Attribution 4.0 International