Machine Learning Method for Changepoint Detection in Short Time Series Data

Simple item page
dc.contributor.authorSmejkalová, Veronikacs
dc.contributor.authorŠomplák, Radovancs
dc.contributor.authorRosecký, Martincs
dc.contributor.authorŠramková, Kristínacs
dc.coverage.issue4cs
dc.coverage.volume5cs
dc.date.accessioned2024-02-20T15:46:08Z
dc.date.available2024-02-20T15:46:08Z
dc.date.issued2023-10-05cs
dc.description.abstractAnalysis of data is crucial in waste management to improve effective planning from both short- and long-term perspectives. Real-world data often presents anomalies, but in the waste management sector, anomaly detection is seldom performed. The main goal and contribution of this paper is a proposal of a complex machine learning framework for changepoint detection in a large number of short time series from waste management. In such a case, it is not possible to use only an expert-based approach due to the time-consuming nature of this process and subjectivity. The proposed framework consists of two steps: (1) outlier detection via outlier test for trend-adjusted data, and (2) changepoints are identified via comparison of linear model parameters. In order to use the proposed method, it is necessary to have a sufficient number of experts’ assessments of the presence of anomalies in time series. The proposed framework is demonstrated on waste management data from the Czech Republic. It is observed that certain waste categories in specific regions frequently exhibit changepoints. On the micro-regional level, approximately 31.1% of time series contain at least one outlier and 16.4% exhibit changepoints. Certain groups of waste are more prone to the occurrence of anomalies. The results indicate that even in the case of aggregated data, anomalies are not rare, and their presence should always be checked.en
dc.formattextcs
dc.format.extent1407-1432cs
dc.format.mimetypeapplication/pdfcs
dc.identifier.citationMachine Learning and Knowledge Extraction. 2023, vol. 5, issue 4, p. 1407-1432.en
dc.identifier.doi10.3390/make5040071cs
dc.identifier.issn2504-4990cs
dc.identifier.orcid0000-0002-6763-2059cs
dc.identifier.orcid0000-0002-5714-4537cs
dc.identifier.orcid0000-0003-0848-7477cs
dc.identifier.other186887cs
dc.identifier.researcheridAAC-8839-2019cs
dc.identifier.researcheridQ-9462-2017cs
dc.identifier.scopus57194697108cs
dc.identifier.scopus55515602000cs
dc.identifier.urihttps://hdl.handle.net/11012/245096
dc.language.isoencs
dc.publisherMDPIcs
dc.relation.ispartofMachine Learning and Knowledge Extractioncs
dc.relation.urihttps://www.mdpi.com/2504-4990/5/4/71cs
dc.rightsCreative Commons Attribution 4.0 Internationalcs
dc.rights.accessopenAccesscs
dc.rights.sherpahttp://www.sherpa.ac.uk/romeo/issn/2504-4990/cs
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/cs
dc.subjectmachine learning for time seriesen
dc.subjectwaste generationen
dc.subjectshort time seriesen
dc.subjectanomaly detectionen
dc.subjectoutlieren
dc.subjectchangepointen
dc.titleMachine Learning Method for Changepoint Detection in Short Time Series Dataen
dc.type.driverarticleen
dc.type.statusPeer-revieweden
dc.type.versionpublishedVersionen
sync.item.dbidVAV-186887en
sync.item.dbtypeVAVen
sync.item.insts2024.02.20 16:46:08en
sync.item.modts2024.02.20 16:13:28en
thesis.grantorVysoké učení technické v Brně. Fakulta strojního inženýrství. Ústav matematikycs
thesis.grantorVysoké učení technické v Brně. Fakulta strojního inženýrství. Ústav procesního inženýrstvícs
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
make0500071.pdf
Size:
3.88 MB
Format:
Adobe Portable Document Format
Description:
file make0500071.pdf
Download
Collections
Ústav matematiky
Ústav procesního inženýrství