Choices and pitfalls concerning mixture-of-experts modeling.

Detalhes bibliográficos
Autor(a) principal: Denise Beatriz Teixeira Pinto
Data de Publicação: 2005
Tipo de documento: Dissertação
Idioma: eng
Título da fonte: Biblioteca Digital de Teses e Dissertações do ITA
Texto Completo: http://www.bd.bibl.ita.br/tde_busca/arquivo.php?codArquivo=197
Resumo: Researchers of different fields often need to learn and represent phenomena through relationships between variables and use them to predict the phenomena behavior under uncertain conditions. However, choosing the "best" model in a modeling exercise is always an arduous task, yet because of the various uncertainties associated with the modeling process.A way of obtaining a better prediction than it would be provided by a single model may be by combining a number of different model structures. Each model is adopted at a given observation with a probability that depends on the values of the explanatory variables for that observation. This is the logic under the mixture-of-experts model (MEM).The general MEM framework specifies that a prediction is made up of a series of predictions from separate models, or experts, each of them weighted by a quantity determined by a so called gating function.However, when building a MEM, many important decisions ought to be made, such as determining the number of clusters in to which the global data is to be sectioned and the clustering method to be adopted; the nature of the gating function applied; the criteria employed to elect the experts and the data set selected for performing validation, for example. Depending on how these decisions are made, different mixtures might be formed, providing different results.In the present study, we investigated the way such decisions affect the performance of the MEM, when this technique employs statistical models, applied to regression problems. The problem under analysis consists of estimating the municipal gross domestic product for the Brazilian states of Pará and Maranhão, as functions of municipal macroeconomic and demographic variables.
id ITA_ab73abbb8a7b62a84bbbf4d62c2b33b9
oai_identifier_str oai:agregador.ibict.br.BDTD_ITA:oai:ita.br:197
network_acronym_str ITA
network_name_str Biblioteca Digital de Teses e Dissertações do ITA
spelling Choices and pitfalls concerning mixture-of-experts modeling.Análise estatísticaModelos matemáticosAnálise de aglomeradosPrediçãoProduto interno brutoIndicadores socioeconômicosInteligência artificialMatemáticaResearchers of different fields often need to learn and represent phenomena through relationships between variables and use them to predict the phenomena behavior under uncertain conditions. However, choosing the "best" model in a modeling exercise is always an arduous task, yet because of the various uncertainties associated with the modeling process.A way of obtaining a better prediction than it would be provided by a single model may be by combining a number of different model structures. Each model is adopted at a given observation with a probability that depends on the values of the explanatory variables for that observation. This is the logic under the mixture-of-experts model (MEM).The general MEM framework specifies that a prediction is made up of a series of predictions from separate models, or experts, each of them weighted by a quantity determined by a so called gating function.However, when building a MEM, many important decisions ought to be made, such as determining the number of clusters in to which the global data is to be sectioned and the clustering method to be adopted; the nature of the gating function applied; the criteria employed to elect the experts and the data set selected for performing validation, for example. Depending on how these decisions are made, different mixtures might be formed, providing different results.In the present study, we investigated the way such decisions affect the performance of the MEM, when this technique employs statistical models, applied to regression problems. The problem under analysis consists of estimating the municipal gross domestic product for the Brazilian states of Pará and Maranhão, as functions of municipal macroeconomic and demographic variables.Instituto Tecnológico de AeronáuticaArmando Zeferino MilioniDenise Beatriz Teixeira Pinto2005-11-11info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesishttp://www.bd.bibl.ita.br/tde_busca/arquivo.php?codArquivo=197reponame:Biblioteca Digital de Teses e Dissertações do ITAinstname:Instituto Tecnológico de Aeronáuticainstacron:ITAenginfo:eu-repo/semantics/openAccessapplication/pdf2019-02-02T14:01:40Zoai:agregador.ibict.br.BDTD_ITA:oai:ita.br:197http://oai.bdtd.ibict.br/requestopendoar:null2020-05-28 19:32:29.906Biblioteca Digital de Teses e Dissertações do ITA - Instituto Tecnológico de Aeronáuticatrue
dc.title.none.fl_str_mv Choices and pitfalls concerning mixture-of-experts modeling.
title Choices and pitfalls concerning mixture-of-experts modeling.
spellingShingle Choices and pitfalls concerning mixture-of-experts modeling.
Denise Beatriz Teixeira Pinto
Análise estatística
Modelos matemáticos
Análise de aglomerados
Predição
Produto interno bruto
Indicadores socioeconômicos
Inteligência artificial
Matemática
title_short Choices and pitfalls concerning mixture-of-experts modeling.
title_full Choices and pitfalls concerning mixture-of-experts modeling.
title_fullStr Choices and pitfalls concerning mixture-of-experts modeling.
title_full_unstemmed Choices and pitfalls concerning mixture-of-experts modeling.
title_sort Choices and pitfalls concerning mixture-of-experts modeling.
author Denise Beatriz Teixeira Pinto
author_facet Denise Beatriz Teixeira Pinto
author_role author
dc.contributor.none.fl_str_mv Armando Zeferino Milioni
dc.contributor.author.fl_str_mv Denise Beatriz Teixeira Pinto
dc.subject.por.fl_str_mv Análise estatística
Modelos matemáticos
Análise de aglomerados
Predição
Produto interno bruto
Indicadores socioeconômicos
Inteligência artificial
Matemática
topic Análise estatística
Modelos matemáticos
Análise de aglomerados
Predição
Produto interno bruto
Indicadores socioeconômicos
Inteligência artificial
Matemática
dc.description.none.fl_txt_mv Researchers of different fields often need to learn and represent phenomena through relationships between variables and use them to predict the phenomena behavior under uncertain conditions. However, choosing the "best" model in a modeling exercise is always an arduous task, yet because of the various uncertainties associated with the modeling process.A way of obtaining a better prediction than it would be provided by a single model may be by combining a number of different model structures. Each model is adopted at a given observation with a probability that depends on the values of the explanatory variables for that observation. This is the logic under the mixture-of-experts model (MEM).The general MEM framework specifies that a prediction is made up of a series of predictions from separate models, or experts, each of them weighted by a quantity determined by a so called gating function.However, when building a MEM, many important decisions ought to be made, such as determining the number of clusters in to which the global data is to be sectioned and the clustering method to be adopted; the nature of the gating function applied; the criteria employed to elect the experts and the data set selected for performing validation, for example. Depending on how these decisions are made, different mixtures might be formed, providing different results.In the present study, we investigated the way such decisions affect the performance of the MEM, when this technique employs statistical models, applied to regression problems. The problem under analysis consists of estimating the municipal gross domestic product for the Brazilian states of Pará and Maranhão, as functions of municipal macroeconomic and demographic variables.
description Researchers of different fields often need to learn and represent phenomena through relationships between variables and use them to predict the phenomena behavior under uncertain conditions. However, choosing the "best" model in a modeling exercise is always an arduous task, yet because of the various uncertainties associated with the modeling process.A way of obtaining a better prediction than it would be provided by a single model may be by combining a number of different model structures. Each model is adopted at a given observation with a probability that depends on the values of the explanatory variables for that observation. This is the logic under the mixture-of-experts model (MEM).The general MEM framework specifies that a prediction is made up of a series of predictions from separate models, or experts, each of them weighted by a quantity determined by a so called gating function.However, when building a MEM, many important decisions ought to be made, such as determining the number of clusters in to which the global data is to be sectioned and the clustering method to be adopted; the nature of the gating function applied; the criteria employed to elect the experts and the data set selected for performing validation, for example. Depending on how these decisions are made, different mixtures might be formed, providing different results.In the present study, we investigated the way such decisions affect the performance of the MEM, when this technique employs statistical models, applied to regression problems. The problem under analysis consists of estimating the municipal gross domestic product for the Brazilian states of Pará and Maranhão, as functions of municipal macroeconomic and demographic variables.
publishDate 2005
dc.date.none.fl_str_mv 2005-11-11
dc.type.driver.fl_str_mv info:eu-repo/semantics/publishedVersion
info:eu-repo/semantics/masterThesis
status_str publishedVersion
format masterThesis
dc.identifier.uri.fl_str_mv http://www.bd.bibl.ita.br/tde_busca/arquivo.php?codArquivo=197
url http://www.bd.bibl.ita.br/tde_busca/arquivo.php?codArquivo=197
dc.language.iso.fl_str_mv eng
language eng
dc.rights.driver.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv Instituto Tecnológico de Aeronáutica
publisher.none.fl_str_mv Instituto Tecnológico de Aeronáutica
dc.source.none.fl_str_mv reponame:Biblioteca Digital de Teses e Dissertações do ITA
instname:Instituto Tecnológico de Aeronáutica
instacron:ITA
reponame_str Biblioteca Digital de Teses e Dissertações do ITA
collection Biblioteca Digital de Teses e Dissertações do ITA
instname_str Instituto Tecnológico de Aeronáutica
instacron_str ITA
institution ITA
repository.name.fl_str_mv Biblioteca Digital de Teses e Dissertações do ITA - Instituto Tecnológico de Aeronáutica
repository.mail.fl_str_mv
subject_por_txtF_mv Análise estatística
Modelos matemáticos
Análise de aglomerados
Predição
Produto interno bruto
Indicadores socioeconômicos
Inteligência artificial
Matemática
_version_ 1706809254476775424