The impact of feature selection methods on online handwritten signature by using clustering-based analysis

Página do item simplificado Estatísticas
dc.contributor.advisorAbreu, Marjory Cristiany da Costa
dc.contributor.advisorIDpt_BR
dc.contributor.advisorLatteshttp://lattes.cnpq.br/2234040548103596pt_BR
dc.contributor.authorMarques, Julliana Caroline Gonçalves de Araújo Silva
dc.contributor.authorIDpt_BR
dc.contributor.authorLatteshttp://lattes.cnpq.br/5554033822360657pt_BR
dc.contributor.referees1Carvalho, Bruno Motta de
dc.contributor.referees1IDpt_BR
dc.contributor.referees1Latteshttp://lattes.cnpq.br/0330924133337698pt_BR
dc.contributor.referees2Souza Neto, Plácido Antônio de
dc.contributor.referees2IDpt_BR
dc.contributor.referees2Latteshttp://lattes.cnpq.br/3641504724164977pt_BR
dc.date.accessioned2021-04-06T19:02:41Z
dc.date.available2021-04-06T19:02:41Z
dc.date.issued2021-01-29
dc.description.resumoHandwritten signature is one of the oldest and most accepted biometric authentication methods for human identity establishment in society. With the popularisation of computers and, consequently, computational biometric authentication systems, the signature was chosen for being one of the biometric traits of an individual that is likely to be relatively unique for every person. However, when dealing with biometric data, including signature data, problems related to high dimensional space, can be generated. Among other issues, irrelevant, redundant data and noise are the most significant, as they result in a decreased of identification accuracy. Thus, it is necessary to reduce the space by selecting the smallest set of features that contain the most discriminative features, increasing the accuracy of the system. In this way, our proposal in this work is to analyse the impact of feature selection on individuals identification accuracy based on the handwritten online signature. For this, we will use two well-known online signature databases: SVC2004 and xLongSignDB. For the feature selection process, we have applied two filter and one wrapper methods. Then, the resulted datasets are evaluated by classification algorithms and validated with a clustering technique. Besides, we have used a statistical test to corroborate our conclusions. Experiments presented satisfactory results when using a smaller number of features which are more representative, showing that we reached an average accuracy of over 98\% for both datasets which were validated with the clustering methods, which achieved an average accuracy over 80\% (SVC2004) and 70\% (xLongSignDB).pt_BR
dc.identifier.citationMARQUES, Julliana Caroline Gonçalves de Araújo Silva. The impact of feature selection methods on online handwritten signature by using clustering-based analysis. 2021. 68f. Dissertação (Mestrado em Sistemas e Computação) - Centro de Ciências Exatas e da Terra, Universidade Federal do Rio Grande do Norte, Natal, 2021.pt_BR
dc.identifier.urihttps://repositorio.ufrn.br/handle/123456789/32052
dc.languagept_BRpt_BR
dc.publisherUniversidade Federal do Rio Grande do Nortept_BR
dc.publisher.countryBrasilpt_BR
dc.publisher.initialsUFRNpt_BR
dc.publisher.programPROGRAMA DE PÓS-GRADUAÇÃO EM SISTEMAS E COMPUTAÇÃOpt_BR
dc.rightsAcesso Abertopt_BR
dc.subjectOnline handwritten signaturept_BR
dc.subjectFeature selectionpt_BR
dc.subjectClusteringpt_BR
dc.subjectSVC2004pt_BR
dc.subjectxLongSignDBpt_BR
dc.titleThe impact of feature selection methods on online handwritten signature by using clustering-based analysispt_BR
dc.typemasterThesispt_BR

Arquivos

Pacote Original

Agora exibindo 1 - 1 de 1
Imagem de Miniatura
Nome:
Impactfeatureselection_Marques_2021.pdf
Tamanho:
14.5 MB
Formato:
Adobe Portable Document Format
Imagem de Miniatura
 Baixar

Coleções

PPGSC - Mestrado em Sistemas e Computação
Repositório Institucional - UFRN Campus Universitário Lagoa NovaCEP 59078-970 Caixa postal 1524 Natal/RN - BrasilUniversidade Federal do Rio Grande do Norte© Copyright 2025. Todos os direitos reservados.
Contato+55 (84) 3342-2260 - R232Setor de Repositórios Digitaisrepositorio@bczm.ufrn.br
DSpaceIBICT
OasisBR
LAReferencia
Customizado pela CAT - BCZM