{"id":81805,"date":"2026-01-28T11:27:03","date_gmt":"2026-01-28T10:27:03","guid":{"rendered":"https:\/\/multi.liora.io\/?p=81805"},"modified":"2026-08-08T12:24:20","modified_gmt":"2026-08-08T11:24:20","slug":"hello-daniel-what-is-data-normalization","status":"publish","type":"post","link":"https:\/\/liora.io\/en\/hello-daniel-what-is-data-normalization","title":{"rendered":"Data normalization: How this concept is related to Data Science?"},"content":{"rendered":"\n

Daniel is the technical support of Liora\u2019s trainings. It is the expert on every subject related to data science. Today, we have managed to get a quick interview with him, so that he can answer a few of our questions about data normalization.<\/strong><\/p>\n\n\n

What is Data Normalization?<\/h2>\n\n\n

Normalization<\/strong>, as it is heard in the data science area, is a very important concept<\/strong> in Data pre-processing, when you need to work on a Machine Learning<\/strong><\/a> project.<\/p>\n\n\n

Two main processes are implied when we talk about normalization: normalization and standard normalization<\/strong>, more commonly known as standardization. Generally, these two processes have the same purpose<\/strong>: to resize numerical variables so that they are comparable on a common scale<\/strong>.<\/p>\n\n\n

In mathematics terms, what do we have?<\/h2>\n\n\n

Let\u2019s consider a numerical variable<\/strong> with n observations, than can be written as followed:<\/p>\n\n\n

\"data-normalization\"<\/figure>\n\n\n

As we have a finite number of real values, we can extract various statistical pieces of information, including min<\/strong>, max mean<\/strong>, and standard deviation<\/strong>. The process of normalization only needs the min and max functions.<\/p>\n\n\n

The purpose here is to bring back all the values of the variable between 0 and 1<\/strong> while keeping some distance between the values.<\/p>\n\n\n

To do that, you\u2019ll use a simple formula:<\/p>\n\n\n

\"Illustration<\/figure>\n\n\n

Regarding the standardization<\/strong>, the transformation is more diffiuclt than easily bringing back the values between 0 and 1. It aims at bringing back the average \u03bc to 0<\/strong> and the standard deviation to 1<\/strong>.<\/p>\n\n\n

Again, the process is not very complicated: if you already know the mean \u03bc<\/strong> and the standard deviation \u03c3<\/strong> of a variable X =\u00a0x1<\/sub>\u00a0x2<\/sub>\u00a0xn<\/sub> you will write the standardized variable<\/strong> as followed:<\/p>\n\n\n

\"Illustration<\/figure>\n\n\n

How do you normalize data concretely?<\/h2>\n\n\n

With Python<\/a><\/strong> it is very simple, many libraries allow it. I will only mention Scikit-learn<\/strong> because it is the most used in Data Science<\/strong>. This library offers functions that perform the desired normalizations in a few simple lines of code.<\/p>\n\n\n

However, it is important to put the use cases<\/strong> in context, because in practice it is not enough to apply a silly normalization to all the Data we have when we already normalized our training data<\/strong>.<\/p>\n\n\n

Why not? The reason is very simple: It is not possible to apply this same transformation to a test sample, or new data<\/strong>.<\/p>\n\n\n

It is obviously possible to center and reduce<\/strong> any sample in the same way, but with an average and standard deviation that will be different from those used on the training set<\/strong>.<\/p>\n\n\n

The results obtained would not be a fair representation of the performance<\/strong> of the model, when applied to new data.<\/p>\n\n\n

So, rather than applying the normalization function directly, it is better to use a Scikit-Learn feature called transformer API<\/strong>, which will allow you to adjust (fit<\/i><\/strong>) a preprocessing step using the training data.<\/p>\n\n\n

So when normalization<\/strong>, for example, is applied to other samples, it will use the same saved average and standard deviations.<\/p>\n\n\n

To create this \u2018adjusted<\/strong>\u2019 preprocessing step, simply use the \u2018StandardScaler<\/i><\/strong>\u2019 function and adjust it using the training data. Finally, to apply it to an array of data afterward, simply apply the following formula: scaler.transform()<\/i><\/strong>.<\/p>\n\n\n

Discover our differents Data Science’s courses<\/a><\/div><\/div>\n\n","protected":false},"excerpt":{"rendered":"

Daniel is the technical support of Liora\u2019s trainings. It is the expert on every subject related to data science. Today, we have managed to get a quick interview with him, so that he can answer a few of our questions about data normalization.<\/p>\n","protected":false},"author":85,"featured_media":30548,"comment_status":"open","ping_status":"open","sticky":false,"template":"elementor_theme","format":"standard","meta":{"_acf_changed":false,"editor_notices":[],"footnotes":""},"categories":[2433],"class_list":["post-81805","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-data-ai"],"acf":[],"_links":{"self":[{"href":"https:\/\/liora.io\/en\/wp-json\/wp\/v2\/posts\/81805","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/liora.io\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/liora.io\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/liora.io\/en\/wp-json\/wp\/v2\/users\/85"}],"replies":[{"embeddable":true,"href":"https:\/\/liora.io\/en\/wp-json\/wp\/v2\/comments?post=81805"}],"version-history":[{"count":5,"href":"https:\/\/liora.io\/en\/wp-json\/wp\/v2\/posts\/81805\/revisions"}],"predecessor-version":[{"id":209785,"href":"https:\/\/liora.io\/en\/wp-json\/wp\/v2\/posts\/81805\/revisions\/209785"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/liora.io\/en\/wp-json\/wp\/v2\/media\/30548"}],"wp:attachment":[{"href":"https:\/\/liora.io\/en\/wp-json\/wp\/v2\/media?parent=81805"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/liora.io\/en\/wp-json\/wp\/v2\/categories?post=81805"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}