Width optimization of RBF kernels for binary classification of support vector machines: a density estimation-based approach

Página do item simplificado
dc.creatorMurilo Menezes
dc.creatorLuiz Carlos Bambirra Torres
dc.creatorAntonio de Padua Braga
dc.date.accessioned2025-05-22T13:38:07Z
dc.date.accessioned2025-09-08T23:50:23Z
dc.date.available2025-05-22T13:38:07Z
dc.date.issued2019
dc.identifier.doihttps://doi.org/10.1016/j.patrec.2019.08.001
dc.identifier.issn0167-8655
dc.identifier.urihttps://hdl.handle.net/1843/82450
dc.languageeng
dc.publisherUniversidade Federal de Minas Gerais
dc.relation.ispartofPattern recognition letters
dc.rightsAcesso Restrito
dc.subjectAprendizagem supervisionada (Aprendizado do computador)
dc.subjectMineração de dados (Computação) - Métodos estatísticos
dc.subject.otherClassification
dc.subject.otherRBF Kernel
dc.subject.otherSupport vector machine
dc.subject.otherDensity estimation
dc.titleWidth optimization of RBF kernels for binary classification of support vector machines: a density estimation-based approach
dc.typeArtigo de periódico
local.citation.epage7
local.citation.issue1
local.citation.spage1
local.citation.volume128
local.description.resumoKernels are often used for modelling non-linear data, developing a main role in models like the SVM. The optimization of its parameters to better fit each dataset is a frequently faced challenge: A bad choice of kernel parameters often implies a poor model. This problem is usually worked out using exhaustive search approaches, such as cross-validation. These methods, however, do not take into account existent information on data arrangement. This paper proposes an alternative approach, based on density estimation. By making use of density estimation methods to analyze the dataset structure, it is proposed a function over the kernel parameters. This function can be used to choose the parameters that best suit the data.
local.publisher.countryBrasil
local.publisher.departmentENG - DEPARTAMENTO DE ENGENHARIA ELETRÔNICA
local.publisher.initialsUFMG
local.url.externahttps://www.sciencedirect.com/science/article/pii/S0167865519302156

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