Exploring early classification strategies of streaming data with delayed attributes
View/ Open
Impact
Scholar |
Other documents of the author: Millán Giraldo, Mónica; Sánchez Garreta, Josep Salvador; Traver Roig, Vicente Javier
Metadata
Show full item recordcomunitat-uji-handle:10234/9
comunitat-uji-handle2:10234/7038
comunitat-uji-handle3:10234/8634
comunitat-uji-handle4:
INVESTIGACIONMetadata
Title
Exploring early classification strategies of streaming data with delayed attributesDate
2009Publisher
Springer Verlag (Germany)ISSN
0302-9743Bibliographic citation
MILLÁN-GIRALDO, Mónica; SÁNCHEZ, J. Salvador; TRAVER, V. Javier. Exploring early classification strategies of streaming data with delayed attributes. En International Conference on Neural Information Processing. Springer, Berlin, Heidelberg, 2009. p. 875-883.MILLÁN-GIRALDO, Mónica; SÁNCHEZ, J. Salvador; TRAVER, V. Javier. Exploring early classification strategies of streaming data with delayed attributes. En International Conference on Neural Information Processing. Springer, Berlin, Heidelberg, 2009. p. 875-883.
Type
info:eu-repo/semantics/articlePublisher version
https://link.springer.com/chapter/10.1007/978-3-642-10677-4_99https://link.springer.com/chapter/10.1007/978-3-642-10677-4_99
Version
info:eu-repo/semantics/acceptedVersionSubject
Abstract
In contrast to traditional machine learning algorithms, where all data are available in batch mode, the new paradigm of streaming data poses additional difficulties, since data samples arrive in a sequence and many ... [+]
In contrast to traditional machine learning algorithms, where all data are available in batch mode, the new paradigm of streaming data poses additional difficulties, since data samples arrive in a sequence and many hard decisions have to be made on-line. The problem addressed here consists of classifying streaming data which not only are unlabeled, but also have a number l of attributes arriving after some time delay. In this context, the main issues are what to do when the unlabeled incomplete samples and, later on, their missing attributes arrive; when and how to classify these incoming samples; and when and how to update the training set. Three different strategies (for l = 1 and constant) are explored and evaluated in terms of the accumulated classification error. The results reveal that the proposed on-line strategies, despite their simplicity, may outperform classifiers using only the original, labeled-and-complete samples as a fixed training set. In other words, learning is possible by properly tapping into the unlabeled, incomplete samples, and their delayed attributes. The many research issues identified include a better understanding of the link between the inherent properties of the data set and the design of the most suitable on-line classification strategy [-]
Is part of
Lecture notes in computer science, 2009, v. 5863Rights
© Springer Verlag (Germany)
http://www.springer.com/open+access/authors+rights?SGWID=0-176704-12-683201-0
info:eu-repo/semantics/openAccess
http://www.springer.com/open+access/authors+rights?SGWID=0-176704-12-683201-0
info:eu-repo/semantics/openAccess
This item appears in the folowing collection(s)
- LSI_Articles [362]