Please use this identifier to cite or link to this item:
http://hdl.handle.net/2445/8524
Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Escalera Guerrero, Sergio | cat |
dc.contributor.author | Tax, David M. J. | cat |
dc.contributor.author | Pujol Vila, Oriol | cat |
dc.contributor.author | Radeva, Petia | cat |
dc.contributor.author | Duin, Robert P. W. | cat |
dc.date.accessioned | 2009-06-05T07:37:03Z | - |
dc.date.available | 2009-06-05T07:37:03Z | - |
dc.date.issued | 2008 | cat |
dc.identifier.issn | 0162-8828 | cat |
dc.identifier.uri | http://hdl.handle.net/2445/8524 | - |
dc.description.abstract | A common way to model multiclass classification problems is by means of Error-Correcting Output Codes (ECOCs). Given a multiclass problem, the ECOC technique designs a code word for each class, where each position of the code identifies the membership of the class for a given binary problem. A classification decision is obtained by assigning the label of the class with the closest code. One of the main requirements of the ECOC design is that the base classifier is capable of splitting each subgroup of classes from each binary problem. However, we cannot guarantee that a linear classifier model convex regions. Furthermore, nonlinear classifiers also fail to manage some type of surfaces. In this paper, we present a novel strategy to model multiclass classification problems using subclass information in the ECOC framework. Complex problems are solved by splitting the original set of classes into subclasses and embedding the binary problems in a problem-dependent ECOC design. Experimental results show that the proposed splitting procedure yields a better performance when the class overlap or the distribution of the training objects conceal the decision boundaries for the base classifier. The results are even more significant when one has a sufficiently large training size. | eng |
dc.format.extent | 14 p. | cat |
dc.format.mimetype | application/pdf | eng |
dc.language.iso | eng | eng |
dc.publisher | IEEE | cat |
dc.relation.isformatof | Reproducció del document publicat a http://dx.doi.org/10.1109/TPAMI.2008.38 | cat |
dc.relation.ispartof | IEEE Transactions on Pattern Analysis and Machine Intelligence, 2008, vol. 30, núm. 6, p. 1041-1054. | cat |
dc.relation.uri | http://dx.doi.org/10.1109/TPAMI.2008.38 | - |
dc.rights | (c) IEEE, 2008 | eng |
dc.source | Articles publicats en revistes (Matemàtiques i Informàtica) | - |
dc.subject.classification | Anàlisi d'error (Matemàtica) | cat |
dc.subject.classification | Categories (Matemàtica) | cat |
dc.subject.other | Error correction codes | eng |
dc.subject.other | Multiclass classification | eng |
dc.subject.other | Subclasses | eng |
dc.subject.other | Embedding of dichotomizers | eng |
dc.title | Subclass problem-dependent design for error-correcting output codes | eng |
dc.type | info:eu-repo/semantics/article | eng |
dc.type | info:eu-repo/semantics/publishedVersion | - |
dc.identifier.idgrec | 557146 | cat |
dc.rights.accessRights | info:eu-repo/semantics/openAccess | - |
Appears in Collections: | Articles publicats en revistes (Matemàtiques i Informàtica) |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
557146.pdf | 6 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.