{"id":9669,"date":"2023-10-11T00:21:40","date_gmt":"2023-10-10T22:21:40","guid":{"rendered":"https:\/\/www.kubicek.ai\/?p=9669"},"modified":"2026-07-28T09:58:43","modified_gmt":"2026-07-28T07:58:43","slug":"machine-learning-how-does-it-actually-work","status":"publish","type":"post","link":"https:\/\/www.kubicek.ai\/en\/machine-learning-how-does-it-actually-work\/","title":{"rendered":"Machine Learning – How Does It Actually Work?"},"content":{"rendered":"\n

Machine learning<\/a> is a term you hear everywhere these days. But how does it actually work?<\/strong> <\/p>\n\n\n

\n
\"\"<\/figure>\n<\/div>\n\n\n

Put simply, the aim is to teach a computer to recognise patterns in data<\/strong> and make decisions<\/strong> on that basis. Instead of writing programs by hand, algorithms are trained on large amounts of data so that they “learn” a given task themselves.<\/p>\n\n\n\n

<\/p>\n\n\n\n

There are a number of machine learning techniques. This article looks at several of them.<\/p>\n\n\n\n

Regression<\/a> \u2013 predicting numerical values<\/h2>\n\n\n\n

Regression models are used when we want to predict a continuous numerical value from input data \u2013 for example, predicting the price of a property from its size, location and so on. The best known regression technique is linear regression, which fits a straight line through the data points so that the prediction error is as small as possible.<\/p>\n\n\n\n

Classification<\/a> \u2013 sorting into classes<\/h2>\n\n\n\n

Unlike regression, classification models try to sort objects into classes or categories \u2013 recognising whether a photo shows a dog or a cat, for instance, or judging from a customer\u2019s behaviour whether they will leave or stay. Common classification techniques include decision trees, neural networks and support vector machines.<\/p>\n\n\n\n

Clustering \u2013 finding similar points<\/h2>\n\n\n\n

Clustering algorithms look for points in a large body of data that resemble one another, then group them into clusters on the basis of that similarity. Clustering is used, for example, to segment customers for targeted marketing, or to organise a large document collection into thematically related groups.<\/p>\n\n\n\n

Dimensionality reduction<\/a> \u2013 simplifying the data<\/h2>\n\n\n\n

We often meet data with a very large number of attributes, which makes it harder to process and analyse. Dimensionality reduction techniques aim to cut the number of attributes while preserving as much information as possible. Principal component analysis, for instance, can replace a large set of original attributes with a small number of new variables.<\/p>\n\n\n\n

Ensemble methods \u2013 strength in numbers<\/h2>\n\n\n\n

Instead of relying on a single model, we build several and let them vote. Rather than one decision tree, for example, we use a set of trees, each trained on a different part of the data, and then combine the results by voting or averaging. This approach often leads to more accurate results.<\/p>\n\n\n\n

Neural networks<\/a> \u2013 inspired by the brain<\/h2>\n\n\n\n

Neural networks imitate the structure of the brain. They consist of simple computational nodes joined by connections, and by setting the weights of those connections appropriately the network can be trained to solve complex problems. One special type of neural network, known as deep learning, has made enormous progress in recent years on tasks such as image and speech recognition.<\/p>\n\n\n\n

Transfer learning<\/a> \u2013 passing knowledge on<\/h2>\n\n\n\n

Instead of training a model from scratch, we can draw on knowledge gained while solving a similar problem. A neural network trained to recognise faces, for instance, can be “retrained” to recognise dogs and cats. Carrying learned features across into a new domain often speeds up training and improves the results.<\/p>\n\n\n\n

<\/p>\n\n\n\n

We have seen several examples of how machine learning works. The key is to find the technique that best matches the type of problem, and to have good data to train the models on. Although the principles are fairly simple, deploying machine learning successfully in practice is very demanding. It is still worth getting into the field, because it opens up fascinating possibilities for both data analysis and practical applications.<\/p>\n","protected":false},"excerpt":{"rendered":"

Machine learning is a term you hear everywhere these days. But how does it actually work? Put simply, the aim is to teach a computer to recognise patterns in data and make decisions on that basis. Instead of writing programs by hand, algorithms are trained on large amounts of data so that they “learn” a […]<\/p>\n","protected":false},"author":1,"featured_media":819,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"_seopress_titles_title":"","_seopress_titles_desc":"","_seopress_robots_index":"","_seopress_robots_follow":"","_seopress_robots_imageindex":"","_seopress_robots_snippet":"","_seopress_robots_primary_cat":"","_seopress_robots_breadcrumbs":"","_seopress_robots_freeze_modified_date":"","_seopress_robots_custom_modified_date":"","_seopress_robots_canonical":"","_seopress_social_fb_title":"","_seopress_social_fb_desc":"","_seopress_social_fb_img":"","_seopress_social_fb_img_attachment_id":0,"_seopress_social_fb_img_width":0,"_seopress_social_fb_img_height":0,"_seopress_social_twitter_title":"","_seopress_social_twitter_desc":"","_seopress_social_twitter_img":"","_seopress_social_twitter_img_attachment_id":0,"_seopress_social_twitter_img_width":0,"_seopress_social_twitter_img_height":0,"_seopress_redirections_value":"","_seopress_redirections_enabled":"","_seopress_redirections_enabled_regex":"","_seopress_redirections_logged_status":"","_seopress_redirections_param":"","_seopress_redirections_type":0,"_seopress_analysis_target_kw":"","_seopress_news_disabled":"","_seopress_video_disabled":"","_seopress_video":[],"_seopress_pro_schemas_manual":[],"_seopress_pro_rich_snippets_disable_all":"","_seopress_pro_rich_snippets_disable":[],"_seopress_pro_schemas":[],"footnotes":""},"categories":[10],"tags":[],"cat_tool":[],"class_list":["post-9669","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-uncategorized"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.kubicek.ai\/en\/wp-json\/wp\/v2\/posts\/9669","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.kubicek.ai\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.kubicek.ai\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.kubicek.ai\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.kubicek.ai\/en\/wp-json\/wp\/v2\/comments?post=9669"}],"version-history":[{"count":2,"href":"https:\/\/www.kubicek.ai\/en\/wp-json\/wp\/v2\/posts\/9669\/revisions"}],"predecessor-version":[{"id":9833,"href":"https:\/\/www.kubicek.ai\/en\/wp-json\/wp\/v2\/posts\/9669\/revisions\/9833"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.kubicek.ai\/en\/wp-json\/wp\/v2\/media\/819"}],"wp:attachment":[{"href":"https:\/\/www.kubicek.ai\/en\/wp-json\/wp\/v2\/media?parent=9669"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.kubicek.ai\/en\/wp-json\/wp\/v2\/categories?post=9669"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.kubicek.ai\/en\/wp-json\/wp\/v2\/tags?post=9669"},{"taxonomy":"cat_tool","embeddable":true,"href":"https:\/\/www.kubicek.ai\/en\/wp-json\/wp\/v2\/cat_tool?post=9669"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}