2026-04-18T03:37:06Zhttps://recercat.cat/oai/request

oai:recercat.cat:2445/1191212025-12-05T09:55:18Zcom_2072_1057col_2072_478917col_2072_478920

Generalized multi-scale stacked sequential learning for multi-class classification Puertas i Prats, Eloi Escalera Guerrero, Sergio Pujol Vila, Oriol Algorismes Aprenentatge Algorithms Learning 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. 2018-01-18T13:43:24Z 2018-01-18T13:43:24Z 2015-04-30 2018-01-18T13:43:24Z info:eu-repo/semantics/article info:eu-repo/semantics/acceptedVersion 1433-7541 https://hdl.handle.net/2445/119121 622017 eng Versió postprint del document publicat a: https://doi.org/10.1007/s10044-013-0333-y Pattern Analysis and Applications, 2015, vol. 18, num. 2, p. 247-261 https://doi.org/10.1007/s10044-013-0333-y (c) Springer Verlag, 2015 info:eu-repo/semantics/openAccess 15 p. application/pdf Springer Verlag Articles publicats en revistes (Matemàtiques i Informàtica)