000845280 001__ 845280
000845280 005__ 20210129233344.0
000845280 0247_ $$2doi$$a10.1016/j.neucom.2017.08.035
000845280 0247_ $$2ISSN$$a0925-2312
000845280 0247_ $$2ISSN$$a1872-8286
000845280 0247_ $$2Handle$$a2128/18233
000845280 0247_ $$2pmid$$apmid:29398782
000845280 0247_ $$2WOS$$aWOS:000418370200032
000845280 037__ $$aFZJ-2018-02561
000845280 082__ $$a610
000845280 1001_ $$00000-0003-1998-8621$$aCollell, Guillem$$b0
000845280 245__ $$aA simple plug-in bagging ensemble based on threshold-moving for classifying binary and multiclass imbalanced data
000845280 260__ $$aAmsterdam$$bElsevier$$c2018
000845280 3367_ $$2DRIVER$$aarticle
000845280 3367_ $$2DataCite$$aOutput Types/Journal article
000845280 3367_ $$0PUB:(DE-HGF)16$$2PUB:(DE-HGF)$$aJournal Article$$bjournal$$mjournal$$s1524488430_26620
000845280 3367_ $$2BibTeX$$aARTICLE
000845280 3367_ $$2ORCID$$aJOURNAL_ARTICLE
000845280 3367_ $$00$$2EndNote$$aJournal Article
000845280 520__ $$aClass imbalance presents a major hurdle in the application of classification methods. A commonly taken approach is to learn ensembles of classifiers using rebalanced data. Examples include bootstrap averaging (bagging) combined with either undersampling or oversampling of the minority class examples. However, rebalancing methods entail asymmetric changes to the examples of different classes, which in turn can introduce their own biases. Furthermore, these methods often require specifying the performance measure of interest a priori, i.e., before learning. An alternative is to employ the threshold moving technique, which applies a threshold to the continuous output of a model, offering the possibility to adapt to a performance measure a posteriori, i.e., a plug-in method. Surprisingly, little attention has been paid to this combination of a bagging ensemble and threshold-moving. In this paper, we study this combination and demonstrate its competitiveness. Contrary to the other resampling methods, we preserve the natural class distribution of the data resulting in well-calibrated posterior probabilities. Additionally, we extend the proposed method to handle multiclass data. We validated our method on binary and multiclass benchmark data sets by using both, decision trees and neural networks as base classifiers. We perform analyses that provide insights into the proposed method.
000845280 536__ $$0G:(DE-HGF)POF3-899$$a899 - ohne Topic (POF3-899)$$cPOF3-899$$fPOF III$$x0
000845280 588__ $$aDataset connected to CrossRef
000845280 7001_ $$0P:(DE-HGF)0$$aPrelec, Drazen$$b1
000845280 7001_ $$0P:(DE-Juel1)172843$$aPatil, Kaustubh$$b2$$eCorresponding author$$ufzj
000845280 773__ $$0PERI:(DE-600)1479006-3$$a10.1016/j.neucom.2017.08.035$$gVol. 275, p. 330 - 340$$p330 - 340$$tNeurocomputing$$v275$$x0925-2312$$y2018
000845280 8564_ $$uhttps://juser.fz-juelich.de/record/845280/files/Guillem18.pdf$$yOpenAccess
000845280 8564_ $$uhttps://juser.fz-juelich.de/record/845280/files/Guillem18.gif?subformat=icon$$xicon$$yOpenAccess
000845280 8564_ $$uhttps://juser.fz-juelich.de/record/845280/files/Guillem18.jpg?subformat=icon-1440$$xicon-1440$$yOpenAccess
000845280 8564_ $$uhttps://juser.fz-juelich.de/record/845280/files/Guillem18.jpg?subformat=icon-180$$xicon-180$$yOpenAccess
000845280 8564_ $$uhttps://juser.fz-juelich.de/record/845280/files/Guillem18.jpg?subformat=icon-640$$xicon-640$$yOpenAccess
000845280 8564_ $$uhttps://juser.fz-juelich.de/record/845280/files/Guillem18.pdf?subformat=pdfa$$xpdfa$$yOpenAccess
000845280 909CO $$ooai:juser.fz-juelich.de:845280$$pdnbdelivery$$pdriver$$pVDB$$popen_access$$popenaire
000845280 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)172843$$aForschungszentrum Jülich$$b2$$kFZJ
000845280 9131_ $$0G:(DE-HGF)POF3-899$$1G:(DE-HGF)POF3-890$$2G:(DE-HGF)POF3-800$$3G:(DE-HGF)POF3$$4G:(DE-HGF)POF$$aDE-HGF$$bProgrammungebundene Forschung$$lohne Programm$$vohne Topic$$x0
000845280 9141_ $$y2018
000845280 915__ $$0StatID:(DE-HGF)0200$$2StatID$$aDBCoverage$$bSCOPUS
000845280 915__ $$0StatID:(DE-HGF)1160$$2StatID$$aDBCoverage$$bCurrent Contents - Engineering, Computing and Technology
000845280 915__ $$0LIC:(DE-HGF)CCBY4$$2HGFVOC$$aCreative Commons Attribution CC BY 4.0
000845280 915__ $$0StatID:(DE-HGF)0600$$2StatID$$aDBCoverage$$bEbsco Academic Search
000845280 915__ $$0StatID:(DE-HGF)0100$$2StatID$$aJCR$$bNEUROCOMPUTING : 2015
000845280 915__ $$0StatID:(DE-HGF)0150$$2StatID$$aDBCoverage$$bWeb of Science Core Collection
000845280 915__ $$0StatID:(DE-HGF)0111$$2StatID$$aWoS$$bScience Citation Index Expanded
000845280 915__ $$0StatID:(DE-HGF)9900$$2StatID$$aIF < 5
000845280 915__ $$0StatID:(DE-HGF)0510$$2StatID$$aOpenAccess
000845280 915__ $$0StatID:(DE-HGF)0030$$2StatID$$aPeer Review$$bASC
000845280 915__ $$0StatID:(DE-HGF)0300$$2StatID$$aDBCoverage$$bMedline
000845280 915__ $$0StatID:(DE-HGF)0199$$2StatID$$aDBCoverage$$bThomson Reuters Master Journal List
000845280 920__ $$lyes
000845280 9201_ $$0I:(DE-Juel1)INM-7-20090406$$kINM-7$$lGehirn & Verhalten$$x0
000845280 980__ $$ajournal
000845280 980__ $$aVDB
000845280 980__ $$aUNRESTRICTED
000845280 980__ $$aI:(DE-Juel1)INM-7-20090406
000845280 9801_ $$aFullTexts