Supervised variational relevance learning, an analytic geometric feature selection with applications to omic datasets

dc.contributorUniversidade de São Paulo (USP)
dc.contributorUniversidade Estadual Paulista (Unesp)
dc.creatorBoareto, Marcelo
dc.creatorCesar, Jonatas
dc.creatorLeite, Vitor Barbanti Pereira [UNESP]
dc.creatorCaticha, Nestor
dc.date2015-10-22T06:48:40Z
dc.date2015-10-22T06:48:40Z
dc.date2015-05-01
dc.date.accessioned2023-09-12T07:02:46Z
dc.date.available2023-09-12T07:02:46Z
dc.identifierhttp://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=6977958
dc.identifierIeee-acm Transactions On Computational Biology And Bioinformatics. Los Alamitos: Ieee Computer Soc, v. 12, n. 3, p. 705-711, 2015.
dc.identifier1545-5963
dc.identifierhttp://hdl.handle.net/11449/129779
dc.identifier10.1109/TCBB.2014.2377750
dc.identifierWOS:000356608100022
dc.identifier0500034174785796
dc.identifier.urihttps://repositorioslatinoamericanos.uchile.cl/handle/2250/8779108
dc.descriptionWe introduce Supervised Variational Relevance Learning (Suvrel), a variational method to determine metric tensors to define distance based similarity in pattern classification, inspired in relevance learning. The variational method is applied to a cost function that penalizes large intraclass distances and favors small interclass distances. We find analytically the metric tensor that minimizes the cost function. Preprocessing the patterns by doing linear transformations using the metric tensor yields a dataset which can be more efficiently classified. We test our methods using publicly available datasets, for some standard classifiers. Among these datasets, two were tested by the MAQC-II project and, even without the use of further preprocessing, our results improve on their performance.
dc.descriptionCenter for the Study of Natural and Artificial Information Processing Systems of the University of Sao Paulo (CNAIPS, Nucleo de Apoio a Pesquisa da Universidade de Sao Paulo)
dc.descriptionFundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)
dc.descriptionConselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)
dc.descriptionInstituto de Física, University of Sao Paulo, Brazil
dc.descriptionIBILCE, Universidade Estadual Paulista, Sao José do Rio Preto, São Paulo,
dc.format705-711
dc.languageeng
dc.publisherIeee Computer Soc
dc.relationIeee-acm Transactions On Computational Biology And Bioinformatics
dc.relation2.428
dc.relation0,649
dc.rightsAcesso restrito
dc.sourceWeb of Science
dc.subjectSuvrel
dc.subjectRelevance Learning
dc.subjectAnalytic metric learning
dc.subjectProteomics
dc.subjectMetabolomics
dc.subjectGenomics
dc.subjectFeature selection
dc.subjectDistance learning
dc.titleSupervised variational relevance learning, an analytic geometric feature selection with applications to omic datasets
dc.typeArtigo


Este ítem pertenece a la siguiente institución