Active feature acquisition on data streams under feature drift

Show simple item record

dc.identifier.uri Beyer, Christian Büttner, Maik Unnikrishnan, Vishnu Schleicher, Miro Ntoutsi, Eirini Spiliopoulou, Myra 2021-05-25T11:49:21Z 2021-05-25T11:49:21Z 2020
dc.identifier.citation Beyer, C.; Büttner, M.; Unnikrishnan, V.; Schleicher, M.; Ntoutsi, E. et al.: Active feature acquisition on data streams under feature drift. In: Annales des Telecommunications/Annals of Telecommunications 75 (2020), S. 597611. DOI:
dc.description.abstract Traditional active learning tries to identify instances for which the acquisition of the label increases model performance under budget constraints. Less research has been devoted to the task of actively acquiring feature values, whereupon both the instance and the feature must be selected intelligently and even less to a scenario where the instances arrive in a stream with feature drift. We propose an active feature acquisition strategy for data streams with feature drift, as well as an active feature acquisition evaluation framework. We also implement a baseline that chooses features randomly and compare the random approach against eight different methods in a scenario where we can acquire at most one feature at the time per instance and where all features are considered to cost the same. Our initial experiments on 9 different data sets, with 7 different degrees of missing features and 8 different budgets show that our developed methods outperform the random acquisition on 7 data sets and have a comparable performance on the remaining two. © 2020, The Author(s). eng
dc.language.iso eng
dc.publisher Berlin : Springer
dc.relation.ispartofseries Annales des Telecommunications/Annals of Telecommunications 75 (2020)
dc.rights CC BY 4.0 Unported
dc.subject Active feature acquisition eng
dc.subject Data streams eng
dc.subject Feature drift eng
dc.subject Budget control eng
dc.subject Petroleum reservoir evaluation eng
dc.subject Active Learning eng
dc.subject Budget constraint eng
dc.subject Evaluation framework eng
dc.subject Feature acquisition eng
dc.subject Feature values eng
dc.subject Missing features eng
dc.subject Model performance eng
dc.subject Data streams eng
dc.subject.ddc 620 | Ingenieurwissenschaften und Maschinenbau ger
dc.title Active feature acquisition on data streams under feature drift
dc.type Article
dc.type Text
dc.relation.essn 1958-9395
dc.relation.issn 0003-4347
dc.bibliographicCitation.volume 75
dc.bibliographicCitation.firstPage 597
dc.bibliographicCitation.lastPage 611
dc.description.version publishedVersion
tib.accessRights frei zug�nglich

Files in this item

This item appears in the following Collection(s):

Show simple item record


Search the repository


My Account

Usage Statistics