{"id":170360,"date":"2023-09-10T08:22:28","date_gmt":"2023-09-10T07:22:28","guid":{"rendered":"https:\/\/liora.io\/en\/?p=170360"},"modified":"2026-02-06T08:59:12","modified_gmt":"2026-02-06T07:59:12","slug":"survival-analysis-beyond-machine-learning","status":"publish","type":"post","link":"https:\/\/liora.io\/en\/survival-analysis-beyond-machine-learning","title":{"rendered":"Survival Analysis: Beyond Machine Learning"},"content":{"rendered":"<p><a href=\"https:\/\/liora.io\/en\/professional-data-science-training-the-best-options\">When starting a Data Science project,<\/a> it&#8217;s crucial to carefully consider the modeling of the problem at hand.<\/p>\n<p>If we aim to increase sales on an e-commerce website, we can work on enhancing conversion rates<a href=\"https:\/\/liora.io\/en\/management-of-unbalanced-classification-problems-i\"> using a classification model,<\/a> determining user visit durations based on their profiles, modeling visitor journeys, attributing visitor arrivals to various marketing channels, and improving website SEO.<\/p>\n<p>There are various mathematical approaches available for these tasks:<\/p>\n<ul>\n<li><a href=\"https:\/\/liora.io\/en\/unlock-your-future-dive-into-machine-learning-engineer-training\">Machine Learning<\/a><\/li>\n<li>survival analysis<\/li>\n<li>Markov chains<\/li>\n<li>Shapley value calculations<\/li>\n<li>PageRank score estimation<\/li>\n<li>and more.<\/li>\n<\/ul>\n<figure>\n\t\t\t\t\t\t\t\t\t\t<img decoding=\"async\" width=\"385\" height=\"368\" src=\"https:\/\/liora.io\/app\/uploads\/sites\/9\/2023\/09\/Survival-Analysis-1.png\" alt=\"\" loading=\"lazy\" srcset=\"https:\/\/liora.io\/app\/uploads\/sites\/9\/2023\/09\/Survival-Analysis-1.png 385w, https:\/\/liora.io\/app\/uploads\/sites\/9\/2023\/09\/Survival-Analysis-1-300x287.png 300w\" sizes=\"(max-width: 385px) 100vw, 385px\"><figcaption><\/figcaption><\/figure>\n<blockquote><p>We can see that <a href=\"https:\/\/liora.io\/en\/machine-learning-engineer-all-about-the-job\">Machine Learning is not the be-all and end-all of the Data Scientist profession.<\/a> It&#8217;s essential to explore other mathematical models based on probability theory, game theory, or graph theory.<\/p><\/blockquote>\n<p>With this in mind, we are going to develop a course on the timely and relevant topic of Survival Analysis.<\/p>\n<figure>\n\t\t\t\t\t\t\t\t\t\t<img decoding=\"async\" width=\"472\" height=\"337\" src=\"https:\/\/liora.io\/app\/uploads\/sites\/9\/2023\/09\/survival-analysis-2.png\" alt=\"\" loading=\"lazy\" srcset=\"https:\/\/liora.io\/app\/uploads\/sites\/9\/2023\/09\/survival-analysis-2.png 472w, https:\/\/liora.io\/app\/uploads\/sites\/9\/2023\/09\/survival-analysis-2-300x214.png 300w\" sizes=\"(max-width: 472px) 100vw, 472px\"><figcaption>Survival function experiencing a slight decrease.<\/figcaption><\/figure>\n<h2>What is survival analysis?<\/h2>\n<p><strong>Survival analysis<\/strong> is a field of statistics that focuses on the lifespan of individuals within a population. Its aim is to estimate the date at which a death event occurs.<\/p>\n<p>However, its scope of application is much broader:<\/p>\n<ol>\n<li><strong>Predictive maintenance:<\/strong> Estimating the <a href=\"https:\/\/liora.io\/en\/data-leakage-definition-and-prevention\">machine failure date<\/a> to intervene in a timely manner.<\/li>\n<li><strong>Churn analysis:<\/strong> Predicting when a customer will unsubscribe from a service.<\/li>\n<li><strong>Credit analysis:<\/strong> Anticipating when a customer may default on payments.<\/li>\n<li><strong>Epidemiology:<\/strong> Forecasting when a patient will recover (in this case, it&#8217;s the virus\/bacteria that &#8220;dies&#8221;).<\/li>\n<\/ol>\n<p>The use of such models dates back to the 1950s in medicine, but some researchers are working on algorithms that combine this <a href=\"https:\/\/liora.io\/en\/data-poisoning-a-threat-to-machine-learning-models\">modeling with Machine Learning techniques.<\/a><\/p>\n<h2>Let&#8217;s take a closer look at the survival function.<\/h2>\n<p>In survival analysis, the goal is to estimate the distribution of a random variable X, which represents the time of an event, such as death. This leads to the introduction of the survival function:<\/p>\n<h5>S(t) = P(X&gt;t)<\/h5>\n<p>and the instantaneous hazard rate:<\/p>\n<p><img decoding=\"async\" width=\"618\" height=\"84\" src=\"https:\/\/liora.io\/app\/uploads\/sites\/9\/2023\/09\/tx-de-risque-instant.png\" alt=\"\" loading=\"lazy\" srcset=\"https:\/\/liora.io\/app\/uploads\/sites\/9\/2023\/09\/tx-de-risque-instant.png 618w, https:\/\/liora.io\/app\/uploads\/sites\/9\/2023\/09\/tx-de-risque-instant-300x41.png 300w\" sizes=\"(max-width: 618px) 100vw, 618px\"><\/p>\n<p>We can estimate these quantities using non-parametric estimators like the Kaplan-Meier estimator, semi-parametric methods like the <strong>Cox model, or parametric models.<\/strong> The latter two types of estimators are especially useful for assessing the influence of explanatory variables on the survival function.<\/p>\n<figure>\n\t\t\t\t\t\t\t\t\t\t<img decoding=\"async\" width=\"800\" height=\"246\" src=\"https:\/\/liora.io\/app\/uploads\/sites\/9\/2023\/09\/survival-analysis-3.png\" alt=\"\" loading=\"lazy\" srcset=\"https:\/\/liora.io\/app\/uploads\/sites\/9\/2023\/09\/survival-analysis-3.png 939w, https:\/\/liora.io\/app\/uploads\/sites\/9\/2023\/09\/survival-analysis-3-300x92.png 300w, https:\/\/liora.io\/app\/uploads\/sites\/9\/2023\/09\/survival-analysis-3-768x236.png 768w\" sizes=\"(max-width: 800px) 100vw, 800px\"><figcaption><\/figcaption><\/figure>\n<p>In survival analysis, the <strong>PySurvival library<\/strong> is highly valuable. It is well-referenced, well-documented, and offers a wide range of useful tools for visualization and performance measurement.<\/p>\n<p>Has this article piqued your interest? Know that a training course on this topic will be starting soon! Feel free to reach out to us for more information!<\/p>\n<p>Are you looking to delve deeper into the possibilities of Machine Learning? Start one of our training courses today!<\/p>\n<p><a href=\"\/en\/courses\/data-ai\/\"><br \/>\nDiscover our courses<br \/>\n<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"<p>When starting a Data Science project, it&#8217;s crucial to carefully consider the modeling of the problem at hand. If we aim to increase sales on an e-commerce website, we can work on enhancing conversion rates using a classification model, determining user visit durations based on their profiles, modeling visitor journeys, attributing visitor arrivals to various [&hellip;]<\/p>\n","protected":false},"author":76,"featured_media":170362,"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-170360","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\/170360","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\/76"}],"replies":[{"embeddable":true,"href":"https:\/\/liora.io\/en\/wp-json\/wp\/v2\/comments?post=170360"}],"version-history":[{"count":1,"href":"https:\/\/liora.io\/en\/wp-json\/wp\/v2\/posts\/170360\/revisions"}],"predecessor-version":[{"id":206368,"href":"https:\/\/liora.io\/en\/wp-json\/wp\/v2\/posts\/170360\/revisions\/206368"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/liora.io\/en\/wp-json\/wp\/v2\/media\/170362"}],"wp:attachment":[{"href":"https:\/\/liora.io\/en\/wp-json\/wp\/v2\/media?parent=170360"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/liora.io\/en\/wp-json\/wp\/v2\/categories?post=170360"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}