Concept learning using one-class classifiers for implicit drift detection in evolving data streams

Simple item page
buir.contributor.authorGözüaçık, Ömer
buir.contributor.authorCan, Fazli
buir.contributor.orcidGözüaçık, Ömer|0000-0002-8200-0017
buir.contributor.orcidCan, Fazli|0000-0003-0016-4278
dc.citation.epage3747en_US
dc.citation.spage3725en_US
dc.citation.volumeNumber54en_US
dc.contributor.authorGözüaçık, Ömer
dc.contributor.authorCan, Fazli
dc.date.accessioned2022-02-04T11:33:36Z
dc.date.available2022-02-04T11:33:36Z
dc.date.issued2021-06
dc.departmentDepartment of Computer Engineeringen_US
dc.description.abstractData stream mining has become an important research area over the past decade due to the increasing amount of data available today. Sources from various domains generate a near-limitless volume of data in temporal order. Such data are referred to as data streams, and are generally nonstationary as the characteristics of data evolves over time. This phe nomenon is called concept drift, and is an issue of great importance in the literature, since it makes models obsolete by decreasing their predictive performance. In the presence of concept drift, it is necessary to adapt to change in data to build more robust and efective classifers. Drift detectors are designed to run jointly with classifcation models, updating them when a signifcant change in data distribution is observed. In this paper, we present an implicit (unsupervised) algorithm called One-Class Drift Detector (OCDD), which uses a one-class learner with a sliding window to detect concept drift. We perform a compre hensive evaluation on mostly recent 17 prevalent concept drift detection methods and an adaptive classifer using 13 datasets. The results show that OCDD outperforms the other methods by producing models with better predictive performance on both real-world and synthetic datasets.en_US
dc.identifier.doi10.1007/s10462-020-09939-xen_US
dc.identifier.eissn1573-7462
dc.identifier.issn0269-2821
dc.identifier.urihttp://hdl.handle.net/11693/77042
dc.language.isoEnglishen_US
dc.publisherSpringeren_US
dc.relation.isversionofhttps://doi.org/10.1007/s10462-020-09939-xen_US
dc.source.titleArtificial Intelligence Reviewen_US
dc.subjectConcept driften_US
dc.subjectData streamen_US
dc.subjectDrift detectionen_US
dc.subjectUnlabeled dataen_US
dc.subjectVerifcation latencyen_US
dc.titleConcept learning using one-class classifiers for implicit drift detection in evolving data streamsen_US
dc.typeArticleen_US

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Concept_learning_using_one‑class_classifers_for_implicit_drift_detection_in_evolving_data_streams.pdf
Size:
1.21 MB
Format:
Adobe Portable Document Format
Description:
Download
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.69 KB
Format:
Item-specific license agreed upon to submission
Description:
Download

Collections

Department of Computer Engineering