{"id":24839,"date":"2019-07-16T00:00:00","date_gmt":"2019-07-16T00:00:00","guid":{"rendered":"https:\/\/www.mailgun.com\/blog\/evitar-pontos-cegos-dados-ausentes-machine-learning\/"},"modified":"2026-09-16T19:09:10","modified_gmt":"2026-09-16T19:09:10","slug":"evitar-pontos-cegos-dados-ausentes-machine-learning","status":"publish","type":"blog","link":"https:\/\/www.mailgun.com\/pt\/blog\/it-and-engineering\/evitar-pontos-cegos-dados-ausentes-machine-learning\/","title":{"rendered":"Como evitar os pontos cegos de dados ausentes com machine learning"},"content":{"rendered":"\n<p class=\"wp-block-paragraph\">You have a project, and you want to apply machine learning to it. You start simple: add one feature, collect data, create a model. You add another feature that\u2019s really useful, but it\u2019s only represented in half of your data points. You want to be smart and use all the data you have (including the one with missing values), but how do you do that?<\/p>\n\r\n    <nav data-content-type=\"longform\" class=\"toc-block longform-spacings px-5 py-6 px-md-6 px-lg-7 py-md-7 bg-light fs-sm rounded-lg\" aria-labelledby=\"toc-title-2088\"><p class=\"h5 m-0\" id=\"toc-title-2088\">\u00cdndice<\/p><div class=\"d-flex mt-3\"><div class=\"w-auto fw-bold text-accent d-flex me-2\">01<\/div><div class=\"d-flex flex-column\"><a class=\"fw-bold scrollme link-body-color text-accent\" href=\"#learning-by-example\">Learning by example<\/a><\/div><\/div><div class=\"d-flex mt-3\"><div class=\"w-auto fw-bold text-accent d-flex me-2\">02<\/div><div class=\"d-flex flex-column\"><a class=\"fw-bold scrollme link-body-color text-accent\" href=\"#the-power-of-data-imputation\">The power of data imputation<\/a><\/div><\/div><div class=\"d-flex mt-3\"><div class=\"w-auto fw-bold text-accent d-flex me-2\">03<\/div><div class=\"d-flex flex-column\"><a class=\"fw-bold scrollme link-body-color text-accent\" href=\"#lessons-learned\">Lessons Learned<\/a><\/div><\/div><div class=\"d-flex mt-3\"><div class=\"w-auto fw-bold text-accent d-flex me-2\">04<\/div><div class=\"d-flex flex-column\"><a class=\"fw-bold scrollme link-body-color text-accent\" href=\"#useful-links\">Useful links<\/a><\/div><\/div><\/nav>\n<figure class=\"wp-block-image size-full\"><img decoding=\"async\" class=\"wp-image-4859\" width=\"500\" height=\"334\" src=\"https:\/\/www.mailgun.com\/wp-content\/uploads\/2025\/07\/sherlock-pose.jpg\" alt=\"Benedict Cumberbatch as Sherlock in solemn pose\" srcset=\"https:\/\/www.mailgun.com\/wp-content\/uploads\/2025\/07\/sherlock-pose.jpg 500w, https:\/\/www.mailgun.com\/wp-content\/uploads\/2025\/07\/sherlock-pose-300x200.jpg 300w\" sizes=\"(max-width: 500px) 100vw, 500px\" \/><\/figure>\n\n<p class=\"wp-block-paragraph\">As a developer, when I run into a problem, I try to google a solution that works. It doesn\u2019t have to be 100% mathematically accurate, but it should make sense. My search led me to\u00a0<a href=\"https:\/\/stackoverflow.com\/questions\/35680426\/missing-value-in-data-analysis\" target=\"_tabs\" rel=\"noopener noreferrer\">Missing Value in Data Analysis<\/a>\u00a0on Stack Overflow.<\/p>\n\n<p class=\"wp-block-paragraph\">Solutions vary from something as simple as filling gaps with mean or most popular values to predicting missing values first. In my case, I introduced a separate binary feature indicating if the value is missing.<\/p>\n\n<p class=\"wp-block-paragraph\">Whenever I doubted a solution, I turned to math. Math is very precise about when something does or does not work, and what the conditions and the trade-offs are. There\u2019s also a lot written about\u00a0<a href=\"https:\/\/en.wikipedia.org\/wiki\/Imputation_(statistics)\" target=\"_tabs\" rel=\"noopener noreferrer\">data imputation<\/a>\u2013 people get doctoral degrees working on this problem!<\/p>\n\n<p class=\"wp-block-paragraph\">But then I came across a different approach (kind of by accident). Instead of trying to impute the data, you can use algorithms that don\u2019t require the data to be imputed. They just work out of the box \u2013 missing values or not. Sounds like a fairy tale, right?<\/p>\n\n<p class=\"wp-block-paragraph\">When I was doing my research, nobody mentioned anything like this to me. I talked to PhDs and people working in the field, and all I heard back was data imputation.<\/p>\n\n<p class=\"wp-block-paragraph\">Then I ran into a\u00a0<a href=\"https:\/\/www.coursera.org\/\" target=\"_tabs\" rel=\"noopener noreferrer\">Coursera<\/a>\u00a0course that went into great detail about decision tree algorithms. The way decision trees work, you start at the root and go left or right with a certain probability. Here\u2019s a decision tree for the survival of passengers on the Titanic:<\/p>\n\n<figure class=\"wp-block-image size-full\"><img decoding=\"async\" class=\"wp-image-4860\" width=\"360\" height=\"340\" src=\"https:\/\/www.mailgun.com\/wp-content\/uploads\/2025\/07\/titanic-probability.png\" alt=\"A decision tree for the outcome of the Titanic survivors\" srcset=\"https:\/\/www.mailgun.com\/wp-content\/uploads\/2025\/07\/titanic-probability.png 360w, https:\/\/www.mailgun.com\/wp-content\/uploads\/2025\/07\/titanic-probability-300x283.png 300w\" sizes=\"(max-width: 360px) 100vw, 360px\" \/><\/figure>\n\n<p class=\"wp-block-paragraph\">The way\u00a0<a href=\"http:\/\/scikit-learn.org\/stable\/modules\/tree.html#tree-algorithms-id3-c4-5-c5-0-and-cart\" target=\"_tabs\" rel=\"noopener noreferrer\">decision tree algorithms<\/a>\u00a0like C4.5, C5.0, and CART account for missing values goes like this:<\/p>\n\n<p class=\"wp-block-paragraph\">Imagine that a feature value is unknown, which means you can\u2019t check the condition and have no way of knowing which branch to follow. One popular approach is to use the most common value. This is essentially the equivalent of picking the most probable branch.<\/p>\n\n<p class=\"wp-block-paragraph\">What tree algorithms do is consider both branches with weights equal to the probability of the branches.<\/p>\n\n<h2 class=\"wp-block-heading\">Learning by example<\/h2>\n\n<p class=\"wp-block-paragraph\">Let\u2019s pull in another example. Here\u2019s a probability tree from a simple\u00a0<a href=\"http:\/\/www.dominatethegmat.com\/2011\/10\/gmat-probability-questions-matrix-problems\/\" target=\"_tabs\" rel=\"noopener noreferrer\">GMAT test<\/a>\u00a0that assumes a sample size of 100 students in a college class:<\/p>\n\n<figure class=\"wp-block-image size-full\"><img decoding=\"async\" class=\"wp-image-4861\" width=\"444\" height=\"343\" src=\"https:\/\/www.mailgun.com\/wp-content\/uploads\/2025\/07\/GMAT-sample-decision-tree.jpg\" alt=\"A GMAT decision tree for college students\" srcset=\"https:\/\/www.mailgun.com\/wp-content\/uploads\/2025\/07\/GMAT-sample-decision-tree.jpg 444w, https:\/\/www.mailgun.com\/wp-content\/uploads\/2025\/07\/GMAT-sample-decision-tree-300x232.jpg 300w\" sizes=\"(max-width: 444px) 100vw, 444px\" \/><\/figure>\n\n<p class=\"wp-block-paragraph\">If you\u2019re male, the probability that you\u2019re single is<\/p>\n\n<p class=\"wp-block-paragraph\"><code>50 \/ 70 = 71%<\/code><\/p>\n\n<p class=\"wp-block-paragraph\">If you\u2019re female, the probability of being single is<\/p>\n\n<p class=\"wp-block-paragraph\"><code>20 \/ 30 = 67%<\/code><\/p>\n\n<p class=\"wp-block-paragraph\">If the gender is unknown, the probability of being single is<\/p>\n\n<p class=\"wp-block-paragraph\"><code>(0.7\u00a0<\/code><i>71%) + (0.3\u00a0<\/i><code>67%) = 70%<\/code><\/p>\n\n<p class=\"wp-block-paragraph\">0.7 because 70 students out of 100 are male. 0.3 because 30 students out of 100 are females.<\/p>\n\n<p class=\"wp-block-paragraph\">By breaking down the overall probability into branch probabilities with weights, we consider all possibilities. In general, I think this is a much better way to overcome missing data and teach our model to generalize future values.<\/p>\n\n<p class=\"wp-block-paragraph\">Unfortunately, libraries that implement these algorithms rarely support missing values. For example,\u00a0<a href=\"http:\/\/scikit-learn.org\/stable\/\" target=\"_tabs\" rel=\"noopener noreferrer\">scikit-learn<\/a>\u00a0library \u2013 the de facto machine learning library for Python \u2013 requires all values to be numeric.<\/p>\n\n<p class=\"wp-block-paragraph\">But there are still good libraries such as\u00a0<a href=\"https:\/\/orange.biolab.si\/\" target=\"_tabs\" rel=\"noopener noreferrer\">Orange<\/a>\u00a0that do support missing values. And as it turns out, the limitation can be overcome.<\/p>\n\n<h2 class=\"wp-block-heading\">The power of data imputation<\/h2>\n\n<p class=\"wp-block-paragraph\">At first, this lack of support for missing values made me feel angry and amused. I mean, seriously, why can\u2019t the very algorithm whose advantage is a built-in support for missing values be used without data imputation?! Come on!<\/p>\n\n<figure class=\"wp-block-image size-full\"><img decoding=\"async\" class=\"wp-image-4862\" width=\"1024\" height=\"576\" src=\"https:\/\/www.mailgun.com\/wp-content\/uploads\/2025\/07\/Dwight-Panic.png\" alt=\"Dwight from &quot;The Office&quot; in a panicked expression\" srcset=\"https:\/\/www.mailgun.com\/wp-content\/uploads\/2025\/07\/Dwight-Panic.png 1024w, https:\/\/www.mailgun.com\/wp-content\/uploads\/2025\/07\/Dwight-Panic-300x169.png 300w, https:\/\/www.mailgun.com\/wp-content\/uploads\/2025\/07\/Dwight-Panic-768x432.png 768w\" sizes=\"(max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n    <div data-content-type=\"longform\" class=\"code-snippet longform-spacings rounded-lg overflow-hidden shadow \" data-count=\"1\">\r\n        <ul class=\"nav nav-buttons code-snippet__tabs longform-except position-relative d-flex gap-2 flex-wrap p-3\" role=\"tablist\">\r\n\r\n                            \r\n                    \r\n                        <li class=\"nav-item\" role=\"presentation\">\r\n                            <button class=\"nav-link active\" data-bs-toggle=\"tab\" data-bs-target=\"#code_0\" type=\"button\" role=\"tab\" aria-controls=\"code_0\" aria-selected=\"true\">\r\n                                PHP                            <\/button>\r\n                        <\/li>\r\n\r\n                    \r\n                            \r\n        <\/ul>\r\n        <div class=\"code-snippet__tab-content tab-content\">\r\n\r\n                            \r\n                    <div class=\"tab-pane show active\" id=\"code_0\" role=\"tabpanel\" aria-labelledby=\"code_0\">\r\n                        <div class=\"code-snippet__code-wrapper\">\r\n                            <div class=\"nav nav-buttons code-snippet__copy-wrapper\">\r\n                                <button type=\"button\" class=\"nav-link code-snippet__copy\" data-clipboard-target=\"#code_0_content\" data-clipboard-label=\"Copiar para a \u00e1rea de transfer\u00eancia\" data-clipboard-success-label=\"Copiado!\" data-clipboard-error-label=\"N\u00e3o foi poss\u00edvel copiar\" data-tooltip=\"Copiar para a \u00e1rea de transfer\u00eancia\" data-tooltip-placement=\"left\" role=\"tooltip\" aria-label=\"Copiar para a \u00e1rea de transfer\u00eancia\">\r\n                                    <svg class=\"code-snippet__copy-icon code-snippet__copy-icon--default\" width=\"24\" height=\"24\" aria-hidden=\"true\" data-url=\"https:\/\/www.mailgun.com\/wp-content\/plugins\/sinch-core\/assets\/icons\/sinch\/file-stack.svg\"><\/svg>\r\n                                    <svg class=\"code-snippet__copy-icon code-snippet__copy-icon--success\" width=\"24\" height=\"24\" aria-hidden=\"true\" data-url=\"https:\/\/www.mailgun.com\/wp-content\/plugins\/sinch-core\/assets\/icons\/sinch\/badge-check.svg\"><\/svg>\r\n                                <\/button>\r\n                            <\/div>\r\n                            <pre class=\"w-100 h-100 m-0 line-numbers language-PHP\">\r\n                                <code id=\"code_0_content\" class=\"language-PHP\">\r\n\r\n                                    import randomrndef impute_gender():rnreturn random.choice([&quot;Male&quot;] * 70 + [&quot;Female&quot;] * 30)\r\n                                <\/code>\r\n                            <\/pre>\r\n                        <\/div>\r\n                    <\/div>\r\n                                    <\/div>\r\n    <\/div>\r\n\n<p class=\"wp-block-paragraph\">And the beauty of data imputation is that it can be applied to any machine learning algorithm, not just decision trees.<\/p>\n\n<p class=\"wp-block-paragraph\">That just blew my mind! An obstacle became a solution, all thanks to the same simple idea!<\/p>\n\n<h2 class=\"wp-block-heading\">Lessons Learned<\/h2>\n\n<p class=\"wp-block-paragraph\">No matter what field you\u2019re working in or how good you are at collecting data, missing values are gonna come up. Maybe you\u2019re working on a credit scoring application. Or maybe you\u2019re trying to predict when email recipients are most likely to open their messages, so you can schedule accordingly. Real tasks tend to have gaps.<\/p>\n\n<p class=\"wp-block-paragraph\">There are so many different ways to think about a problem like missing values, and depending on your case, the answers can be different. But in the heart of a complex solution often lies a simple idea.<\/p>\n\n<h2 class=\"wp-block-heading\">Useful links<\/h2>\n\n<ul class=\"wp-block-list\">\n<li>Wiki article on <a href=\"https:\/\/en.wikipedia.org\/wiki\/Imputation_(statistics)\" target=\"_tabs\" rel=\"noopener noreferrer\">data imputation<\/a><\/li>\n\n\n\n<li>Some common ways to <a href=\"https:\/\/stackoverflow.com\/questions\/35680426\/missing-value-in-data-analysis\" target=\"_tabs\" rel=\"noopener noreferrer\">input missing values<\/a>\u00a0<\/li>\n\n\n\n<li>Quora article on how <a href=\"https:\/\/www.quora.com\/In-simple-language-how-does-C4-5-deal-with-missing-values\" target=\"_tabs\" rel=\"noopener noreferrer\">decision trees handle missing values<\/a>\u00a0<\/li>\n\n\n\n<li><a href=\"https:\/\/orange.biolab.si\/\" target=\"_tabs\" rel=\"noopener noreferrer\">Orange<\/a> &#8211; commonly used machine learning library in Python that supports missing values<\/li>\n\n\n\n<li><a href=\"https:\/\/www.mailgun.com\/pt\/blog\/machine-learning-for-everyday-tasks\" target=\"_tabs\" rel=\"noopener noreferrer\">Testing your model<\/a> to withstand real-world tasks<\/li>\n<\/ul>\n\n<p class=\"wp-block-paragraph\">Happy machine learning! How do you deal with missing values? Tell me down below in the comments\u2026<\/p>\n\r\n    <aside data-content-type=\"longform\"  class=\"banner-block longform-spacings rounded-lg shadow-lg px-5 py-6 px-md-6 py-md-7 p-lg-7 bg-light\" data-theme=\"dark\" aria-labelledby=\"banner-block-block_7328caceac85c5f35c56ccae16b8860f\">\r\n                    <p class=\"text-uppercase section-caption text-body-color\">\r\n                Sign Up            <\/p>\r\n                            <p class=\"h4 text-accent fw-bold\" id=\"banner-block-block_7328caceac85c5f35c56ccae16b8860f\">\r\n                It&#8217;s easy to get started. And it&#8217;s free.            <\/p>\r\n                            <div class=\"mb-0 text-body-color\">\r\n                                    <div class=\"content-body mb-0 text-body-color\">See what you can accomplish with the world\u2019s best email delivery platform.<\/div>\r\n                            <\/div>\r\n        \r\n        <div class=\"text-start mt-5\"><a href=\"https:\/\/signup.mailgun.com\/new\/signup\"  class=\"btn btn-secondary\" >Get started<\/a><\/div>\r\n    <\/aside>\r\n\r\n","protected":false},"excerpt":{"rendered":"<p>Voc\u00ea tem um projeto e quer aplicar machine learning nele. Voc\u00ea come\u00e7a com o b\u00e1sico: adiciona um atributo, coleta dados, cria um modelo. Voc\u00ea adiciona outro atributo muito \u00fatil, mas ele est\u00e1 presente em apenas metade dos seus pontos de dados. Voc\u00ea quer agir com intelig\u00eancia e usar todos os dados que tem (incluindo aqueles com valores ausentes), mas como fazer isso?<\/p>\n","protected":false},"author":47,"featured_media":4843,"menu_order":0,"template":"","meta":{"_acf_changed":false,"footnotes":""},"blog_category":[23],"class_list":["post-24839","blog","type-blog","status-publish","has-post-thumbnail","hentry","blog_category-it-and-engineering"],"acf":[],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v28.4 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>Prever valores ausentes com imputa\u00e7\u00e3o de dados em modelos de machine learning - Transactional Email API Service For Developers | Mailgun<\/title>\n<meta name=\"description\" content=\"Contorne e treine seu modelo para inferir valores ausentes atrav\u00e9s de \u00e1rvores de decis\u00e3o e algoritmos de machine learning para resolver tarefas cotidianas.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/www.mailgun.com\/pt\/blog\/it-and-engineering\/evitar-pontos-cegos-dados-ausentes-machine-learning\/\" \/>\n<meta property=\"og:locale\" content=\"pt_BR\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Prever valores ausentes com imputa\u00e7\u00e3o de dados em modelos de machine learning - Transactional Email API Service For Developers | Mailgun\" \/>\n<meta property=\"og:description\" content=\"Contorne e treine seu modelo para inferir valores ausentes atrav\u00e9s de \u00e1rvores de decis\u00e3o e algoritmos de machine learning para resolver tarefas cotidianas.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.mailgun.com\/pt\/blog\/it-and-engineering\/evitar-pontos-cegos-dados-ausentes-machine-learning\/\" \/>\n<meta property=\"og:site_name\" content=\"Transactional Email API Service For Developers | Mailgun\" \/>\n<meta property=\"article:modified_time\" content=\"2026-09-16T19:09:10+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/www.mailgun.com\/wp-content\/uploads\/2025\/07\/MG-Blog-IT-Engineering.png\" \/>\n\t<meta property=\"og:image:width\" content=\"720\" \/>\n\t<meta property=\"og:image:height\" content=\"448\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/png\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data1\" content=\"5 minutos\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/www.mailgun.com\\\/pt\\\/blog\\\/it-and-engineering\\\/evitar-pontos-cegos-dados-ausentes-machine-learning\\\/\",\"url\":\"https:\\\/\\\/www.mailgun.com\\\/pt\\\/blog\\\/it-and-engineering\\\/evitar-pontos-cegos-dados-ausentes-machine-learning\\\/\",\"name\":\"Prever valores ausentes com imputa\u00e7\u00e3o de dados em modelos de machine learning - Transactional Email API Service For Developers | Mailgun\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/www.mailgun.com\\\/pt\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/www.mailgun.com\\\/pt\\\/blog\\\/it-and-engineering\\\/evitar-pontos-cegos-dados-ausentes-machine-learning\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/www.mailgun.com\\\/pt\\\/blog\\\/it-and-engineering\\\/evitar-pontos-cegos-dados-ausentes-machine-learning\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/www.mailgun.com\\\/wp-content\\\/uploads\\\/2025\\\/07\\\/MG-Blog-IT-Engineering.png\",\"datePublished\":\"2019-07-16T00:00:00+00:00\",\"dateModified\":\"2026-09-16T19:09:10+00:00\",\"description\":\"Contorne e treine seu modelo para inferir valores ausentes atrav\u00e9s de \u00e1rvores de decis\u00e3o e algoritmos de machine learning para resolver tarefas cotidianas.\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/www.mailgun.com\\\/pt\\\/blog\\\/it-and-engineering\\\/evitar-pontos-cegos-dados-ausentes-machine-learning\\\/#breadcrumb\"},\"inLanguage\":\"pt-BR\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/www.mailgun.com\\\/pt\\\/blog\\\/it-and-engineering\\\/evitar-pontos-cegos-dados-ausentes-machine-learning\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"pt-BR\",\"@id\":\"https:\\\/\\\/www.mailgun.com\\\/pt\\\/blog\\\/it-and-engineering\\\/evitar-pontos-cegos-dados-ausentes-machine-learning\\\/#primaryimage\",\"url\":\"https:\\\/\\\/www.mailgun.com\\\/wp-content\\\/uploads\\\/2025\\\/07\\\/MG-Blog-IT-Engineering.png\",\"contentUrl\":\"https:\\\/\\\/www.mailgun.com\\\/wp-content\\\/uploads\\\/2025\\\/07\\\/MG-Blog-IT-Engineering.png\",\"width\":720,\"height\":448},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/www.mailgun.com\\\/pt\\\/blog\\\/it-and-engineering\\\/evitar-pontos-cegos-dados-ausentes-machine-learning\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/www.mailgun.com\\\/pt\\\/inicio\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Blog\",\"item\":\"https:\\\/\\\/www.mailgun.com\\\/pt\\\/blog\\\/\"},{\"@type\":\"ListItem\",\"position\":3,\"name\":\"Como evitar os pontos cegos de dados ausentes com machine learning\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/www.mailgun.com\\\/pt\\\/#website\",\"url\":\"https:\\\/\\\/www.mailgun.com\\\/pt\\\/\",\"name\":\"Transactional Email API Service For Developers | Mailgun\",\"description\":\"Poderosas APIs de e-mail transacional que permitem a voc\u00ea enviar, receber e rastrear e-mails, criadas com os desenvolvedores em mente. Saiba mais hoje mesmo!\",\"publisher\":{\"@id\":\"https:\\\/\\\/www.mailgun.com\\\/pt\\\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/www.mailgun.com\\\/pt\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"pt-BR\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/www.mailgun.com\\\/pt\\\/#organization\",\"name\":\"Transactional Email API Service For Developers | Mailgun\",\"url\":\"https:\\\/\\\/www.mailgun.com\\\/pt\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"pt-BR\",\"@id\":\"https:\\\/\\\/www.mailgun.com\\\/pt\\\/#\\\/schema\\\/logo\\\/image\\\/\",\"url\":\"https:\\\/\\\/www.mailgun.com\\\/wp-content\\\/uploads\\\/2025\\\/06\\\/cropped-android-chrome-512x512-1.png\",\"contentUrl\":\"https:\\\/\\\/www.mailgun.com\\\/wp-content\\\/uploads\\\/2025\\\/06\\\/cropped-android-chrome-512x512-1.png\",\"width\":512,\"height\":512,\"caption\":\"Transactional Email API Service For Developers | Mailgun\"},\"image\":{\"@id\":\"https:\\\/\\\/www.mailgun.com\\\/pt\\\/#\\\/schema\\\/logo\\\/image\\\/\"}}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Prever valores ausentes com imputa\u00e7\u00e3o de dados em modelos de machine learning - Transactional Email API Service For Developers | Mailgun","description":"Contorne e treine seu modelo para inferir valores ausentes atrav\u00e9s de \u00e1rvores de decis\u00e3o e algoritmos de machine learning para resolver tarefas cotidianas.","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/www.mailgun.com\/pt\/blog\/it-and-engineering\/evitar-pontos-cegos-dados-ausentes-machine-learning\/","og_locale":"pt_BR","og_type":"article","og_title":"Prever valores ausentes com imputa\u00e7\u00e3o de dados em modelos de machine learning - Transactional Email API Service For Developers | Mailgun","og_description":"Contorne e treine seu modelo para inferir valores ausentes atrav\u00e9s de \u00e1rvores de decis\u00e3o e algoritmos de machine learning para resolver tarefas cotidianas.","og_url":"https:\/\/www.mailgun.com\/pt\/blog\/it-and-engineering\/evitar-pontos-cegos-dados-ausentes-machine-learning\/","og_site_name":"Transactional Email API Service For Developers | Mailgun","article_modified_time":"2026-09-16T19:09:10+00:00","og_image":[{"width":720,"height":448,"url":"https:\/\/www.mailgun.com\/wp-content\/uploads\/2025\/07\/MG-Blog-IT-Engineering.png","type":"image\/png"}],"twitter_card":"summary_large_image","twitter_misc":{"Est. reading time":"5 minutos"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/www.mailgun.com\/pt\/blog\/it-and-engineering\/evitar-pontos-cegos-dados-ausentes-machine-learning\/","url":"https:\/\/www.mailgun.com\/pt\/blog\/it-and-engineering\/evitar-pontos-cegos-dados-ausentes-machine-learning\/","name":"Prever valores ausentes com imputa\u00e7\u00e3o de dados em modelos de machine learning - Transactional Email API Service For Developers | Mailgun","isPartOf":{"@id":"https:\/\/www.mailgun.com\/pt\/#website"},"primaryImageOfPage":{"@id":"https:\/\/www.mailgun.com\/pt\/blog\/it-and-engineering\/evitar-pontos-cegos-dados-ausentes-machine-learning\/#primaryimage"},"image":{"@id":"https:\/\/www.mailgun.com\/pt\/blog\/it-and-engineering\/evitar-pontos-cegos-dados-ausentes-machine-learning\/#primaryimage"},"thumbnailUrl":"https:\/\/www.mailgun.com\/wp-content\/uploads\/2025\/07\/MG-Blog-IT-Engineering.png","datePublished":"2019-07-16T00:00:00+00:00","dateModified":"2026-09-16T19:09:10+00:00","description":"Contorne e treine seu modelo para inferir valores ausentes atrav\u00e9s de \u00e1rvores de decis\u00e3o e algoritmos de machine learning para resolver tarefas cotidianas.","breadcrumb":{"@id":"https:\/\/www.mailgun.com\/pt\/blog\/it-and-engineering\/evitar-pontos-cegos-dados-ausentes-machine-learning\/#breadcrumb"},"inLanguage":"pt-BR","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.mailgun.com\/pt\/blog\/it-and-engineering\/evitar-pontos-cegos-dados-ausentes-machine-learning\/"]}]},{"@type":"ImageObject","inLanguage":"pt-BR","@id":"https:\/\/www.mailgun.com\/pt\/blog\/it-and-engineering\/evitar-pontos-cegos-dados-ausentes-machine-learning\/#primaryimage","url":"https:\/\/www.mailgun.com\/wp-content\/uploads\/2025\/07\/MG-Blog-IT-Engineering.png","contentUrl":"https:\/\/www.mailgun.com\/wp-content\/uploads\/2025\/07\/MG-Blog-IT-Engineering.png","width":720,"height":448},{"@type":"BreadcrumbList","@id":"https:\/\/www.mailgun.com\/pt\/blog\/it-and-engineering\/evitar-pontos-cegos-dados-ausentes-machine-learning\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/www.mailgun.com\/pt\/inicio\/"},{"@type":"ListItem","position":2,"name":"Blog","item":"https:\/\/www.mailgun.com\/pt\/blog\/"},{"@type":"ListItem","position":3,"name":"Como evitar os pontos cegos de dados ausentes com machine learning"}]},{"@type":"WebSite","@id":"https:\/\/www.mailgun.com\/pt\/#website","url":"https:\/\/www.mailgun.com\/pt\/","name":"Transactional Email API Service For Developers | Mailgun","description":"Poderosas APIs de e-mail transacional que permitem a voc\u00ea enviar, receber e rastrear e-mails, criadas com os desenvolvedores em mente. Saiba mais hoje mesmo!","publisher":{"@id":"https:\/\/www.mailgun.com\/pt\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.mailgun.com\/pt\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"pt-BR"},{"@type":"Organization","@id":"https:\/\/www.mailgun.com\/pt\/#organization","name":"Transactional Email API Service For Developers | Mailgun","url":"https:\/\/www.mailgun.com\/pt\/","logo":{"@type":"ImageObject","inLanguage":"pt-BR","@id":"https:\/\/www.mailgun.com\/pt\/#\/schema\/logo\/image\/","url":"https:\/\/www.mailgun.com\/wp-content\/uploads\/2025\/06\/cropped-android-chrome-512x512-1.png","contentUrl":"https:\/\/www.mailgun.com\/wp-content\/uploads\/2025\/06\/cropped-android-chrome-512x512-1.png","width":512,"height":512,"caption":"Transactional Email API Service For Developers | Mailgun"},"image":{"@id":"https:\/\/www.mailgun.com\/pt\/#\/schema\/logo\/image\/"}}]}},"_links":{"self":[{"href":"https:\/\/www.mailgun.com\/pt\/wp-json\/wp\/v2\/blog\/24839","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.mailgun.com\/pt\/wp-json\/wp\/v2\/blog"}],"about":[{"href":"https:\/\/www.mailgun.com\/pt\/wp-json\/wp\/v2\/types\/blog"}],"author":[{"embeddable":true,"href":"https:\/\/www.mailgun.com\/pt\/wp-json\/wp\/v2\/users\/47"}],"version-history":[{"count":3,"href":"https:\/\/www.mailgun.com\/pt\/wp-json\/wp\/v2\/blog\/24839\/revisions"}],"predecessor-version":[{"id":24991,"href":"https:\/\/www.mailgun.com\/pt\/wp-json\/wp\/v2\/blog\/24839\/revisions\/24991"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.mailgun.com\/pt\/wp-json\/wp\/v2\/media\/4843"}],"wp:attachment":[{"href":"https:\/\/www.mailgun.com\/pt\/wp-json\/wp\/v2\/media?parent=24839"}],"wp:term":[{"taxonomy":"blog_category","embeddable":true,"href":"https:\/\/www.mailgun.com\/pt\/wp-json\/wp\/v2\/blog_category?post=24839"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}