Please use this identifier to cite or link to this item:
Full metadata record
DC FieldValueLanguage
dc.contributor.authorHoi, Steven C. H.en
dc.contributor.authorWang, Jialei.en
dc.contributor.authorZhao, Peilin.en
dc.description.abstractBoth cost-sensitive classification and online learning have been studied extensively in data mining and machine learning communities, respectively. It is a bit surprising that there was very limited comprehensive study for addressing an important intersecting problem, that is, cost-sensitive online classification. In this paper, we formally study this problem, and propose a new framework for cost-sensitive online classification by exploiting the idea of online gradient descent techniques. Based on the framework, we propose a family of cost-sensitive online classification algorithms, which are designed to directly optimize two well-known cost-sensitive measures: (i) maximization of weighted sum of sensitivity and specificity, and (ii) minimization of weighted misclassification cost. We analyze the theoretical bounds of the cost-sensitive measures made by the proposed algorithms, and extensively examine their empirical performance on a variety of cost-sensitive online classification tasks.en
dc.subjectDRNTU::Engineering::Computer science and engineeringen
dc.titleCost-sensitive online classificationen
dc.typeConference Paperen
dc.contributor.schoolSchool of Computer Engineeringen
dc.contributor.conferenceIEEE International Conference on Data Mining (12th : 2012 : Brussels, Belgium)en
item.fulltextNo Fulltext-
Appears in Collections:SCSE Conference Papers

Google ScholarTM



Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.