Please use this identifier to cite or link to this item:
Title: Generalized multi-scale stacked sequential learning for multi-class classification
Author: Puertas i Prats, Eloi
Escalera Guerrero, Sergio
Pujol Vila, Oriol
Keywords: Algorismes
Issue Date: 30-Apr-2015
Publisher: Springer Verlag
Abstract: In many classification problems, neighbor data labels have inherent sequential relationships. Sequential learning algorithms take benefit of these relationships in order to improve generalization. In this paper, we revise the multi-scale sequential learning approach (MSSL) for applying it in the multi-class case (MMSSL). We introduce the error-correcting output codesframework in the MSSL classifiers and propose a formulation for calculating confidence maps from the margins of the base classifiers. In addition, we propose a MMSSL compression approach which reduces the number of features in the extended data set without a loss in performance. The proposed methods are tested on several databases, showing significant performance improvement compared to classical approaches.
Note: Versió postprint del document publicat a:
It is part of: Pattern Analysis and Applications, 2015, vol. 18, num. 2, p. 247-261
Related resource:
ISSN: 1433-7541
Appears in Collections:Articles publicats en revistes (Matemàtiques i Informàtica)

Files in This Item:
File Description SizeFormat 
622017.pdf2.21 MBAdobe PDFView/Open

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.