{"id":158,"date":"2020-07-18T16:59:26","date_gmt":"2020-07-18T19:59:26","guid":{"rendered":"https:\/\/eic.cefet-rj.br\/~jsoares\/?p=158"},"modified":"2020-08-16T09:45:26","modified_gmt":"2020-08-16T12:45:26","slug":"158","status":"publish","type":"post","link":"https:\/\/eic.cefet-rj.br\/~jsoares\/2020\/07\/18\/158\/","title":{"rendered":"Defesa de disserta\u00e7\u00e3o (14\/08\/2020): Thiago da Silva Pereira"},"content":{"rendered":"<header class=\"entry-header\">\n<h2 class=\"entry-title\">Defesa de disserta\u00e7\u00e3o (14\/08\/2020): Thiago da Silva Pereira<\/h2>\n<\/header>\n<div class=\"entry-content\">\n<div class=\"pf-content\">\n<p><strong>Discente:<\/strong> Thiago da Silva Pereira<\/p>\n<p><strong>T\u00edtulo:<\/strong> Imputa\u00e7\u00e3o de dados <em>hot-deck<\/em>: uma compara\u00e7\u00e3o entre comit\u00eas de regress\u00e3o (<em>Hot-Deck Data Imputation: a comparison among ensemble methods<\/em>)<\/p>\n<p><strong>Orientadores:<\/strong>\u00a0 Jorge de Abreu Soares (orientador) e Eduardo Bezerra da Silva (CEFET\/RJ) (coorientador).<\/p>\n<p><strong>Banca:<\/strong> Jorge de Abreu Soares (presidente), Eduardo Bezerra da Silva (CEFET\/RJ), Diego Nunes Brand\u00e3o (CEFET\/RJ) e Ronaldo Ribeiro Goldschmidt (IME)<\/p>\n<p><strong>Dia\/hora:<\/strong> 14 de agosto de 2020, \u00e0s 15h.<\/p>\n<p><strong>Sala remota:<\/strong> <a href=\"http:\/\/meet.google.com\/mtr-vmkq-wrw\">meet.google.com\/mtr-vmkq-wrw<\/a><\/p>\n<p><strong>Resumo:<\/strong><\/p>\n<p>O problema da aus\u00eancia de dados em conjuntos de dados \u00e9 relevante e dentre as maneiras de se lidar com este problema, a substitui\u00e7\u00e3o do valor ausente por outro (tamb\u00e9m chamada de imputa\u00e7\u00e3o de dados) produz um ganho substancial no aprendizado de m\u00e1quina subsequente. Diversos algoritmos de aprendizado de m\u00e1quina s\u00e3o estudados para a imputa\u00e7\u00e3o de dados, por\u00e9m poucos estudos utilizam m\u00e9todos ensemble para a gera\u00e7\u00e3o do dado a ser imputado. Este trabalho pretende realizar uma compara\u00e7\u00e3o entre diversos m\u00e9todos ensemble (<em>bagging<\/em>, <em>adaboost<\/em>, <em>gradientboost<\/em> e <em>stacked generalization<\/em>) para imputa\u00e7\u00e3o de dados, executando as simula\u00e7\u00f5es em tr\u00eas conjuntos de dados diferentes (<em>AIDS Deaths &#8211; National Health and Family Planning Commission of China<\/em>, <em>Breast Cancer<\/em> e <em>Photometric redshift estimation<\/em>) com 10%, 20% e 30% de dados ausentes, combinando a execu\u00e7\u00e3o das tarefas de agrupamento e redu\u00e7\u00e3o de dimensionalidade com percentuais de redu\u00e7\u00e3o de 10%, 20% e 30% antes da imputa\u00e7\u00e3o.<\/p>\n<\/div>\n<p><strong>Abstract:<\/strong><\/p>\n<p>Preprocessing data faces an important question related to deal with missing data. A possible solution to resolve this challenge is hot-deck imputation. This technique has two steps: group similar records and performs imputation. Selecting the best algorithm for imputation is a challenge. Several machine learning algorithms are studied for data imputation, however few studies compare ensemble methods for the imputation stage. This study proposes a solution based on hot-deck imputation comparing four ensemble regressors: Bagging, Adaboost, Gradientboost, and Stacked Generalization. To ascertain effectiveness, we have used three datasets, varying missing rates from 10% to 30%. Results measuring the precision of imputed data by both techniques indicate that the Gradientboost reveals better precision in reasonable processing time.<\/p>\n<p>&nbsp;<\/p>\n<\/div>\n","protected":false},"excerpt":{"rendered":"<p>Defesa de disserta\u00e7\u00e3o (14\/08\/2020): Thiago da Silva Pereira Discente: Thiago da Silva Pereira T\u00edtulo: Imputa\u00e7\u00e3o de dados hot-deck: uma compara\u00e7\u00e3o entre comit\u00eas de regress\u00e3o (Hot-Deck Data Imputation: a comparison among ensemble methods) Orientadores:\u00a0 Jorge de Abreu Soares (orientador) e Eduardo Bezerra da Silva (CEFET\/RJ) (coorientador). Banca: Jorge de Abreu Soares (presidente), Eduardo Bezerra da Silva [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"aside","meta":{"footnotes":""},"categories":[1],"tags":[],"class_list":["post-158","post","type-post","status-publish","format-aside","hentry","category-defesas-orientandos","post_format-post-format-aside"],"_links":{"self":[{"href":"https:\/\/eic.cefet-rj.br\/~jsoares\/wp-json\/wp\/v2\/posts\/158","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/eic.cefet-rj.br\/~jsoares\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/eic.cefet-rj.br\/~jsoares\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/eic.cefet-rj.br\/~jsoares\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/eic.cefet-rj.br\/~jsoares\/wp-json\/wp\/v2\/comments?post=158"}],"version-history":[{"count":8,"href":"https:\/\/eic.cefet-rj.br\/~jsoares\/wp-json\/wp\/v2\/posts\/158\/revisions"}],"predecessor-version":[{"id":208,"href":"https:\/\/eic.cefet-rj.br\/~jsoares\/wp-json\/wp\/v2\/posts\/158\/revisions\/208"}],"wp:attachment":[{"href":"https:\/\/eic.cefet-rj.br\/~jsoares\/wp-json\/wp\/v2\/media?parent=158"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/eic.cefet-rj.br\/~jsoares\/wp-json\/wp\/v2\/categories?post=158"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/eic.cefet-rj.br\/~jsoares\/wp-json\/wp\/v2\/tags?post=158"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}