Speeding up optimum-path forest training by path-cost propagation

dc.contributorUniversidade Estadual Paulista (Unesp)
dc.contributorUniversidade Estadual de Campinas (UNICAMP)
dc.contributorUniversity of Fortaleza
dc.contributorUniversity of Porto
dc.date.accessioned2014-05-27T11:27:22Z
dc.date.available2014-05-27T11:27:22Z
dc.date.created2014-05-27T11:27:22Z
dc.date.issued2012-12-01
dc.identifierProceedings - International Conference on Pattern Recognition, p. 1233-1236.
dc.identifier1051-4651
dc.identifierhttp://hdl.handle.net/11449/73946
dc.identifier2-s2.0-84874569486
dc.identifier9039182932747194
dc.description.abstractIn this paper we present an optimization of the Optimum-Path Forest classifier training procedure, which is based on a theoretical relationship between minimum spanning forest and optimum-path forest for a specific path-cost function. Experiments on public datasets have shown that the proposed approach can obtain similar accuracy to the traditional one but with faster data training. © 2012 ICPR Org Committee.
dc.languageeng
dc.relationProceedings - International Conference on Pattern Recognition
dc.relation0,307
dc.rightsAcesso aberto
dc.sourceScopus
dc.subjectMinimum spanning forests
dc.subjectOptimum-path forests
dc.subjectSoftware engineering
dc.subjectPattern recognition
dc.titleSpeeding up optimum-path forest training by path-cost propagation
dc.typeActas de congresos


Este ítem pertenece a la siguiente institución