elib
DLR-Header
DLR-Logo -> http://www.dlr.de
DLR Portal Home | Imprint | Privacy Policy | Accessibility | Contact | Deutsch
Fontsize: [-] Text [+]

Increasing Explainability in Time Series Classification by Functional Decomposition

de Graaff, Thies and Wild, Michael and Werner, Tino and Möhlmann, Eike and Seibt, Stefan and Ebrecht, Benjamin (2024) Increasing Explainability in Time Series Classification by Functional Decomposition. In: Explainable Artificial Intelligence, 2156, pp. 125-144. Springer Nature Switzerland. The 2nd World Conference on eXplainable Artificial Intelligence, 2024-07-17 - 2024-07-19, Valletta, Malta. doi: 10.1007/978-3-031-63803-9_7. ISBN 978-3-031-63803-9.

[img] PDF - Only accessible within DLR
876kB
[img] PDF
1MB

Official URL: https://link.springer.com/chapter/10.1007/978-3-031-63803-9_7

Abstract

In this work, we develop a generic methodology for time series classification in order to increase explainability and trustworthiness of the predictions. We achieve this by dividing the whole time series into sub-sequences with a sliding window approach, transforming and classifying each chunk using paradigms from functional decomposition, and then aggregating the results to a final class. Visual as well as dataset-based explanations will be derived, enabling the investigation of errors. We demonstrate our concepts on a case study in the railway domain, where the train type is to be inferred based on the signal of a commonly used axle counting sensor. While outperforming LSTM- and FCN-based end-to-end models regarding the classification accuracy, our method further reveals interesting insights into causes of misclassifications.

Item URL in elib:https://elib.dlr.de/205812/
Document Type:Conference or Workshop Item (Speech)
Title:Increasing Explainability in Time Series Classification by Functional Decomposition
Authors:
AuthorsInstitution or Email of AuthorsAuthor's ORCID iDORCID Put Code
de Graaff, Thiesthies.degraaff (at) dlr.dehttps://orcid.org/0009-0006-5918-9524UNSPECIFIED
Wild, Michaelmichael.wild (at) dlr.dehttps://orcid.org/0009-0009-4120-5524165745014
Werner, Tinotino.werner (at) dlr.dehttps://orcid.org/0000-0002-3512-8667UNSPECIFIED
Möhlmann, EikeEike.Moehlmann (at) dlr.dehttps://orcid.org/0000-0003-3815-6353UNSPECIFIED
Seibt, Stefanstefan.seibt (at) tu-berlin.deUNSPECIFIEDUNSPECIFIED
Ebrecht, Benjaminbenjamin.ebrecht (at) tu-berlin.deUNSPECIFIEDUNSPECIFIED
Date:10 July 2024
Journal or Publication Title:Explainable Artificial Intelligence
Refereed publication:Yes
Open Access:No
Gold Open Access:No
In SCOPUS:Yes
In ISI Web of Science:No
Volume:2156
DOI:10.1007/978-3-031-63803-9_7
Page Range:pp. 125-144
Editors:
EditorsEmailEditor's ORCID iDORCID Put Code
Longo, LucaTechnological University DublinUNSPECIFIEDUNSPECIFIED
Lapuschkin, SebastianFraunhofer Institute for TelecommunicationsUNSPECIFIEDUNSPECIFIED
Seifert, ChristinUniversity of MarburgUNSPECIFIEDUNSPECIFIED
Publisher:Springer Nature Switzerland
Series Name:Communications in Computer and Information Science
ISBN:978-3-031-63803-9
Status:Published
Keywords:Explainable AI, Neuro-symbolic AI, Time Series Classification
Event Title:The 2nd World Conference on eXplainable Artificial Intelligence
Event Location:Valletta, Malta
Event Type:international Conference
Event Start Date:17 July 2024
Event End Date:19 July 2024
HGF - Research field:Aeronautics, Space and Transport
HGF - Program:Transport
HGF - Program Themes:Rail Transport
DLR - Research area:Transport
DLR - Program:V SC Schienenverkehr
DLR - Research theme (Project):V - INTRA - Infrastruktur und Transformation
Location: Oldenburg
Institutes and Institutions:Institute of Systems Engineering for Future Mobility > Systems Theory and Design
Deposited By: de Graaff, Thies
Deposited On:19 Aug 2024 07:03
Last Modified:12 Sep 2024 11:05

Repository Staff Only: item control page

Browse
Search
Help & Contact
Information
OpenAIRE Validator logo electronic library is running on EPrints
Website and database design: Copyright © German Aerospace Center (DLR). All rights reserved.