Signal propagation in Bayesian networks and its relationship with intrinsically multivariate predictive variables

Página do item simplificado
dc.contributor.authorMartins, David C.
dc.contributor.authorOliveira, Evaldo A. de [UNIFESP]
dc.contributor.authorBraga-Neto, Ulisses M.
dc.contributor.authorHashimoto, Ronaldo F.
dc.contributor.authorCesar, Roberto M.
dc.contributor.institutionFed Univ ABC
dc.contributor.institutionUniversidade Federal de São Paulo (UNIFESP)
dc.contributor.institutionUniversidade de São Paulo (USP)
dc.contributor.institutionBrazilian Bioethanol Sci & Technol Lab
dc.contributor.institutionTexas A&M Univ
dc.date.accessioned2016-01-24T14:31:24Z
dc.date.available2016-01-24T14:31:24Z
dc.date.issued2013-03-10
dc.description.abstractA set of predictor variables is said to be intrinsically multivariate predictive (IMP) for a target variable if all properly contained subsets of the predictor set are poor predictors of the. target but the full set predicts the target with great accuracy. in a previous article, the main properties of IMP Boolean variables have been analytically described, including the introduction of the IMP score, a metric based on the coefficient of determination (CoD) as a measure of predictiveness with respect to the target variable. It was shown that the IMP score depends on four main properties: logic of connection, predictive power, covariance between predictors and marginal predictor probabilities (biases). This paper extends that work to a broader context, in an attempt to characterize properties of discrete Bayesian networks that contribute to the presence of variables (network nodes) with high IMP scores. We have found that there is a relationship between the IMP score of a node and its territory size, i.e., its position along a pathway with one source: nodes far from the source display larger IMP scores than those closer to the source, and longer pathways display larger maximum IMP scores. This appears to be a consequence of the fact that nodes with small territory have larger probability of having highly covariate predictors, which leads to smaller IMP scores. in addition, a larger number of XOR and NXOR predictive logic relationships has positive influence over the maximum IMP score found in the pathway. This work presents analytical results based on a simple structure network and an analysis involving random networks constructed by computational simulations. Finally, results from a real Bayesian network application are provided. (C) 2012 Elsevier Inc. All rights reserved.en
dc.description.affiliationFed Univ ABC, Ctr Math Computat & Cognit, BR-09210170 Santo Andre, SP, Brazil
dc.description.affiliationUniversidade Federal de São Paulo, Dept Earth & Exact Sci, BR-09972270 Diadema, SP, Brazil
dc.description.affiliationUniv São Paulo, Inst Math & Stat, BR-05508090 São Paulo, Brazil
dc.description.affiliationBrazilian Bioethanol Sci & Technol Lab, BR-13083970 Campinas, SP, Brazil
dc.description.affiliationTexas A&M Univ, Genom Signal Proc Lab, College Stn, TX 77843 USA
dc.description.affiliationUnifespUniversidade Federal de São Paulo, Dept Earth & Exact Sci, BR-09972270 Diadema, SP, Brazil
dc.description.sourceWeb of Science
dc.description.sponsorshipFundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)
dc.description.sponsorshipCoordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)
dc.description.sponsorshipConselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)
dc.description.sponsorshipMicrosoft Research
dc.description.sponsorshipU.S. National Science Foundation, through NSF
dc.description.sponsorshipIDU.S. National Science Foundation, through NSF: CCF-0845407
dc.format.extent18-34
dc.identifierhttp://dx.doi.org/10.1016/j.ins.2012.10.027
dc.identifier.citationInformation Sciences. New York: Elsevier B.V., v. 225, p. 18-34, 2013.
dc.identifier.doi10.1016/j.ins.2012.10.027
dc.identifier.issn0020-0255
dc.identifier.urihttp://repositorio.unifesp.br/handle/11600/36084
dc.identifier.wosWOS:000314084700002
dc.language.isoeng
dc.publisherElsevier B.V.
dc.relation.ispartofInformation Sciences
dc.rightsinfo:eu-repo/semantics/restrictedAccess
dc.rights.licensehttp://www.elsevier.com/about/open-access/open-access-policies/article-posting-policy
dc.subjectBayesian networken
dc.subjectFeature selectionen
dc.subjectIntrinsically multivariate predictionen
dc.titleSignal propagation in Bayesian networks and its relationship with intrinsically multivariate predictive variablesen
dc.typeinfo:eu-repo/semantics/article
Arquivos
Coleções
ICAQF - Artigos