Multi-objective decision in machine learning

Página do item simplificado
dc.creatorTalles Henriquede de Medeiros
dc.creatorHonovan Paz Rocha
dc.creatorFrank Sill Torres
dc.creatorRicardo Hiroshi Caldeira Takahashi
dc.creatorAntônio Pádua Braga
dc.date.accessioned2025-03-26T14:50:34Z
dc.date.accessioned2025-09-08T23:07:07Z
dc.date.available2025-03-26T14:50:34Z
dc.date.issued2016
dc.identifier.doi10.1007/s40313-016-0295-6
dc.identifier.issn2195-3880
dc.identifier.urihttps://hdl.handle.net/1843/80940
dc.languageeng
dc.publisherUniversidade Federal de Minas Gerais
dc.relation.ispartofJournal of Control, Automation and Electrical Systems
dc.rightsAcesso Restrito
dc.subjectInteligência artificial
dc.subjectAprendizado do computador
dc.subjectSistemas especialistas (Computação)
dc.subjectRecuperação da informação
dc.subject.otherknowledge is effectively applied to model a decision mechanism in the Pareto-optimal set
dc.subject.otherthe method MOBJ-htnn was developed with the intention of enabling the use of the knowledge about the inherent imprecision of the data acquisition process
dc.titleMulti-objective decision in machine learning
dc.typeArtigo de periódico
local.citation.epage227
local.citation.issue2
local.citation.spage217
local.citation.volume28
local.description.resumoThis work presents a novel approach for decision-making for multi-objective binary classification problems. The purpose of the decision process is to select within a set of Pareto-optimal solutions, one model that minimizes the structural risk (generalization error). This new approach utilizes a kind of prior knowledge that, if available, allows the selection of a model that better represents the problem in question. Prior knowledge about the imprecisions of the collected data enables the identification of the region of equivalent solutions within the set of Pareto-optimal solutions. Results for binary classification problems with sets of synthetic and real data indicate equal or better performance in terms of decision efficiency compared to similar approaches.
local.publisher.countryBrasil
local.publisher.departmentENG - DEPARTAMENTO DE ENGENHARIA ELETRÔNICA
local.publisher.departmentICX - DEPARTAMENTO DE MATEMÁTICA
local.publisher.initialsUFMG
local.url.externahttps://link.springer.com/article/10.1007/s40313-016-0295-6

Arquivos

Licença do pacote

Agora exibindo 1 - 1 de 1
Imagem de Miniatura
Nome:
License.txt
Tamanho:
1.99 KB
Formato:
Plain Text
Descrição:
Baixar

Coleções

Artigo de Periódico

Logo UFMG