Active Learning From Stream Data Using Optimal Weight Classifier Ensemble

UTSePress Research/Manakin Repository

Search UTSePress Research


Advanced Search

Browse

My Account

Show simple item record

dc.contributor.author Zhu, Xingquan en_US
dc.contributor.author Lin, Xuemin en_US
dc.contributor.author Shi, Yuquan en_US
dc.contributor.author Zhang, Peng en_US
dc.contributor.editor en_US
dc.date.accessioned 2012-02-02T04:23:38Z
dc.date.available 2012-02-02T04:23:38Z
dc.date.issued 2010 en_US
dc.identifier 2010002553 en_US
dc.identifier.citation Zhu Xingquan et al. 2010, 'Active Learning From Stream Data Using Optimal Weight Classifier Ensemble', Ieee-Inst Electrical Electronics Engineers Inc, vol. 40, no. 6, pp. 1607-1621. en_US
dc.identifier.issn 1083-4419 en_US
dc.identifier.other C1 en_US
dc.identifier.uri http://hdl.handle.net/10453/14521
dc.description.abstract In this paper, we propose a new research problem on active learning from data streams, where data volumes grow continuously, and labeling all data is considered expensive and impractical. The objective is to label a small portion of stream data from which a model is derived to predict future instances as accurately as possible. To tackle the technical challenges raised by the dynamic nature of the stream data, i.e., increasing data volumes and evolving decision concepts, we propose a classifierensemble- based active learning framework that selectively labels instances from data streams to build a classifier ensemble. We argue that a classifier ensemblea??s variance directly corresponds to its error rate, and reducing a classifier ensemblea??s variance is equivalent to improving its prediction accuracy. Because of this, one should label instances toward theminimization of the variance of the underlying classifier ensemble. Accordingly, we introduce a minimum-variance (MV) principle to guide the instance labeling process for data streams. In addition, we derive an optimal-weight calculationmethod to determine the weight values for the classifier ensemble. The MV principle and the optimal weighting module are combined to build an active learning framework for data streams. Experimental results on synthetic and real-world data demonstrate the performance of the proposed work in comparison with other approaches. en_US
dc.language en_US
dc.publisher IEEE-Inst Electrical Electronics Engineers Inc en_US
dc.relation.isbasedon http://dx.doi.org/10.1109/TSMCB.2010.2042445 en_US
dc.title Active Learning From Stream Data Using Optimal Weight Classifier Ensemble en_US
dc.parent Ieee Transactions On Systems Man And Cybernetics Part B-Cybernetics en_US
dc.journal.volume 40 en_US
dc.journal.number 6 en_US
dc.publocation Piscataway en_US
dc.identifier.startpage 1607 en_US
dc.identifier.endpage 1621 en_US
dc.cauo.name FEIT.School of Elec, Mech and Mechatronic Systems en_US
dc.conference Verified OK en_US
dc.for 010200 en_US
dc.personcode 107283 en_US
dc.personcode 0000059262 en_US
dc.personcode 0000022938 en_US
dc.personcode 120662 en_US
dc.percentage 50 en_US
dc.classification.name Applied Mathematics en_US
dc.classification.type FOR-08 en_US
dc.edition en_US
dc.custom en_US
dc.date.activity en_US
dc.location.activity ISI:000284364400016 en_US
dc.description.keywords Active learning, classifier ensemble, stream data en_US


Files in this item

This item appears in the following Collection(s)

Show simple item record