[BibTeX] [RIS]
Efficient prediction algorithms for binary decomposition techniques
Type of publication: Article
Citation: jf:DAMI
Journal: Data Mining and Knowledge Discovery
Volume: 24
Number: 1
Year: 2012
Pages: 40-77
Publisher: Springer Netherlands
ISSN: 1384-5810
URL: /publications/papers/dami12.pdf
DOI: 10.1007/s10618-011-0219-9
Abstract: Binary decomposition methods transform multiclass learning problems into a series of two-class learning problems that can be solved with simpler learning algorithms. As the number of such binary learning problems often grows super-linearly with the number of classes, we need efficient methods for computing the predictions. In this article, we discuss an efficient algorithm that queries only a dynamically determined subset of the trained classifiers, but still predicts the same classes that would have been predicted if all classifiers had been queried. The algorithm is first derived for the simple case of pairwise classification, and then generalized to arbitrary pairwise decompositions of the learning problem in the form of ternary error-correcting output codes under a variety of different code designs and decoding strategies.
Keywords: aggregation, binary decomposition, efficient decoding, efficient voting, multiclass classification, pairwise classification, ternary ECOC
Authors Park, Sang-Hyeun
Fürnkranz, Johannes
Topics