{"id":2934,"date":"2020-05-25T08:47:32","date_gmt":"2020-05-25T08:47:32","guid":{"rendered":"https:\/\/datascientest.com\/?p=2934"},"modified":"2023-11-09T14:14:33","modified_gmt":"2023-11-09T14:14:33","slug":"performance-and-interpretability-in-machine-learning","status":"publish","type":"post","link":"https:\/\/datascientest.com\/performance-and-interpretability-in-machine-learning","title":{"rendered":"Machine Learning : Performance and interpretability"},"content":{"rendered":"\t\t<div data-elementor-type=\"wp-post\" data-elementor-id=\"2934\" class=\"elementor elementor-2934\" data-elementor-post-type=\"post\">\n\t\t\t\t\t\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-166103c elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"166103c\" data-element_type=\"section\">\n\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"elementor-column elementor-col-100 elementor-top-column elementor-element elementor-element-803be8e\" data-id=\"803be8e\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-fb3353c elementor-widget elementor-widget-heading\" data-id=\"fb3353c\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<style>\/*! elementor - v3.15.0 - 02-08-2023 *\/\n.elementor-heading-title{padding:0;margin:0;line-height:1}.elementor-widget-heading .elementor-heading-title[class*=elementor-size-]>a{color:inherit;font-size:inherit;line-height:inherit}.elementor-widget-heading .elementor-heading-title.elementor-size-small{font-size:15px}.elementor-widget-heading .elementor-heading-title.elementor-size-medium{font-size:19px}.elementor-widget-heading .elementor-heading-title.elementor-size-large{font-size:29px}.elementor-widget-heading .elementor-heading-title.elementor-size-xl{font-size:39px}.elementor-widget-heading .elementor-heading-title.elementor-size-xxl{font-size:59px}<\/style><h2 class=\"elementor-heading-title elementor-size-default\">Performance and interpretability: A necessary trade-off?<\/h2>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-d2d791b elementor-widget elementor-widget-text-editor\" data-id=\"d2d791b\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<style>\/*! elementor - v3.15.0 - 02-08-2023 *\/\n.elementor-widget-text-editor.elementor-drop-cap-view-stacked .elementor-drop-cap{background-color:#69727d;color:#fff}.elementor-widget-text-editor.elementor-drop-cap-view-framed .elementor-drop-cap{color:#69727d;border:3px solid;background-color:transparent}.elementor-widget-text-editor:not(.elementor-drop-cap-view-default) .elementor-drop-cap{margin-top:8px}.elementor-widget-text-editor:not(.elementor-drop-cap-view-default) .elementor-drop-cap-letter{width:1em;height:1em}.elementor-widget-text-editor .elementor-drop-cap{float:left;text-align:center;line-height:1;font-size:50px}.elementor-widget-text-editor .elementor-drop-cap-letter{display:inline-block}<\/style>\t\t\t\t<h4><span style=\"font-weight: normal;\">Over the last decade, an increasing number of companies have embraced the digital transformation by incorporating Artificial Intelligence methods in the way they conceive their products and define their processes. <\/span>Gathering, analyzing, and harnessing the data have been considered as essential drivers for business growth in an increasingly number of fields<span style=\"font-weight: normal;\">.<\/span><\/h4>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-866eef9 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"866eef9\" data-element_type=\"section\">\n\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"elementor-column elementor-col-100 elementor-top-column elementor-element elementor-element-deabead\" data-id=\"deabead\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-a1ca378 elementor-widget elementor-widget-text-editor\" data-id=\"a1ca378\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\tHowever, as these tasks have become common practices for most companies of our digital era, many concerns are nowadays raised concerning the application of AI for specific purposes or in particular sectors.\nLet\u2019s take a example in our day-to-day-life to illustrate this statement. Imagine your application for a loan is refused based on some scoring algorithms newly implemented by your bank. Although these algorithms might be efficient in the way they assess people\u2019s risk profile, their lack of transparency put bank advisors in a difficult situation where they are unable to justify the bank decision.<\/span>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-b30288f elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"b30288f\" data-element_type=\"section\">\n\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"elementor-column elementor-col-100 elementor-top-column elementor-element elementor-element-d8ffcc7\" data-id=\"d8ffcc7\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-1f0b6f5 elementor-widget elementor-widget-text-editor\" data-id=\"1f0b6f5\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\tThe opaqueness of Artificial Intelligence is one of the major obstacles of its adoption by most of companies and organisations. It represents a permanent challenge for data scientists as they have to ensure a high accuracy of their model while maintaining a sufficient level of comprehensibility.<\/span>\n<blockquote>In this context, understanding the importance of model interpretability to solve concrete problems is the first step towards successful Data Science projects.<\/blockquote>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-bfc262d elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"bfc262d\" data-element_type=\"section\">\n\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"elementor-column elementor-col-100 elementor-top-column elementor-element elementor-element-5dbf625\" data-id=\"5dbf625\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-3f13e99 elementor-widget elementor-widget-heading\" data-id=\"3f13e99\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<h2 class=\"elementor-heading-title elementor-size-default\">Explicability vs. Interpretability\n<\/h2>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-46a980a elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"46a980a\" data-element_type=\"section\">\n\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"elementor-column elementor-col-100 elementor-top-column elementor-element elementor-element-203f345\" data-id=\"203f345\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-2d46f35 elementor-widget elementor-widget-text-editor\" data-id=\"2d46f35\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\tLet\u2019s start with essential definitions.<br><b>Explicability.<\/b> An algorithmic decision is said to be explainable if it can be explicitly accounted for on the basis of known data and characteristics of the situation. In other words, if it is possible to relate the values taken by certain variables (the characteristics) and their consequences on the prediction, for example of a score, and thus on the decision.\n<blockquote>To use the example of the bank loan, the model can be considered as explainable if it explicitly indicates the relationship between the values taken by the variables (age, family situation, salary, etc.) and the final score.<\/span><\/blockquote>\n<b>Interpretability.<\/b> An algorithmic decision is said to be interpretable if it is possible to <\/span>identify <span style=\"font-weight: 400;\">the characteristics or <b>variables that participate most in the decision<\/b>, or even to <b>quantify their importance<\/b><\/h5>\n<blockquote>For instance, the model used for loan scoring is interpretable if it is able to measure the relative importance of the variables used (age, family situation, salary, etc.) in the determination of the final score.<\/blockquote>\nNote that an explainable decision is interpretable.\nNow we are clear about the context, how does it apply in practice?\u00a0\n<br>The graph below provides a graphical illustration for case of the credit loan application\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-67ceea1 elementor-widget elementor-widget-image\" data-id=\"67ceea1\" data-element_type=\"widget\" data-widget_type=\"image.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<style>\/*! elementor - v3.15.0 - 02-08-2023 *\/\n.elementor-widget-image{text-align:center}.elementor-widget-image a{display:inline-block}.elementor-widget-image a img[src$=\".svg\"]{width:48px}.elementor-widget-image img{vertical-align:middle;display:inline-block}<\/style>\t\t\t\t\t\t\t\t\t<figure class=\"wp-caption\">\n\t\t\t\t\t\t\t\t\t\t<img decoding=\"async\" width=\"800\" height=\"527\" src=\"https:\/\/datascientest.com\/wp-content\/uploads\/2020\/05\/Capture-d\u2019e\u0301cran-2020-05-20-a\u0300-13.02.29-1024x674.png\" class=\"attachment-large size-large wp-image-2813\" alt=\"Explicabilit\u00e9 vs interpr\u00e9tabilit\u00e9\" srcset=\"https:\/\/datascientest.com\/wp-content\/uploads\/2020\/05\/Capture-d\u2019e\u0301cran-2020-05-20-a\u0300-13.02.29-1024x674.png 1024w, https:\/\/datascientest.com\/wp-content\/uploads\/2020\/05\/Capture-d\u2019e\u0301cran-2020-05-20-a\u0300-13.02.29-300x197.png 300w, https:\/\/datascientest.com\/wp-content\/uploads\/2020\/05\/Capture-d\u2019e\u0301cran-2020-05-20-a\u0300-13.02.29-768x505.png 768w, https:\/\/datascientest.com\/wp-content\/uploads\/2020\/05\/Capture-d\u2019e\u0301cran-2020-05-20-a\u0300-13.02.29.png 1356w\" sizes=\"(max-width: 800px) 100vw, 800px\" \/>\t\t\t\t\t\t\t\t\t\t\t<figcaption class=\"widget-image-caption wp-caption-text\">Figure \u2013 Illustration for the the case of credit loan application, Adapted from [3]<\/figcaption>\n\t\t\t\t\t\t\t\t\t\t<\/figure>\n\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-bcd49cd elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"bcd49cd\" data-element_type=\"section\">\n\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"elementor-column elementor-col-100 elementor-top-column elementor-element elementor-element-b3d2a5a\" data-id=\"b3d2a5a\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-10e50de elementor-widget elementor-widget-heading\" data-id=\"10e50de\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<h2 class=\"elementor-heading-title elementor-size-default\">Why is it important? \n\n<\/h2>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-44017d6 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"44017d6\" data-element_type=\"section\">\n\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"elementor-column elementor-col-100 elementor-top-column elementor-element elementor-element-5e02422\" data-id=\"5e02422\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-0a02af4 elementor-widget elementor-widget-image\" data-id=\"0a02af4\" data-element_type=\"widget\" data-widget_type=\"image.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<img decoding=\"async\" width=\"800\" height=\"486\" src=\"https:\/\/datascientest.com\/wp-content\/uploads\/2020\/05\/Capture-d\u2019e\u0301cran-2020-05-20-a\u0300-13.02.37.png\" class=\"attachment-large size-large wp-image-2814\" alt=\"Explicabilit\u00e9 vs interpr\u00e9tabilit\u00e9\" srcset=\"https:\/\/datascientest.com\/wp-content\/uploads\/2020\/05\/Capture-d\u2019e\u0301cran-2020-05-20-a\u0300-13.02.37.png 896w, https:\/\/datascientest.com\/wp-content\/uploads\/2020\/05\/Capture-d\u2019e\u0301cran-2020-05-20-a\u0300-13.02.37-300x182.png 300w, https:\/\/datascientest.com\/wp-content\/uploads\/2020\/05\/Capture-d\u2019e\u0301cran-2020-05-20-a\u0300-13.02.37-768x466.png 768w\" sizes=\"(max-width: 800px) 100vw, 800px\" \/>\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-f214d16 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"f214d16\" data-element_type=\"section\">\n\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"elementor-column elementor-col-100 elementor-top-column elementor-element elementor-element-7d51e9b\" data-id=\"7d51e9b\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-c1a2437 elementor-widget elementor-widget-heading\" data-id=\"c1a2437\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<h2 class=\"elementor-heading-title elementor-size-default\">A Classification of Machine Learning Algorithms<\/h2>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-36efeee elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"36efeee\" data-element_type=\"section\">\n\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"elementor-column elementor-col-100 elementor-top-column elementor-element elementor-element-e0e5b2b\" data-id=\"e0e5b2b\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-d8ab792 elementor-widget elementor-widget-text-editor\" data-id=\"d8ab792\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\tSo, how Machine Learning algorithms can be divided between the two categories?\n<br>Broadly speaking, machine learning algorithms can be structured in two groups according to whether they lead, by construction, to an explicit model (white box) or a black box.\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-ecb0d99 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"ecb0d99\" data-element_type=\"section\">\n\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"elementor-column elementor-col-100 elementor-top-column elementor-element elementor-element-a3756a3\" data-id=\"a3756a3\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap\">\n\t\t\t\t\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-e4de78a elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"e4de78a\" data-element_type=\"section\">\n\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"elementor-column elementor-col-100 elementor-top-column elementor-element elementor-element-4e972bf\" data-id=\"4e972bf\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-83c9a67 elementor-widget elementor-widget-text-editor\" data-id=\"83c9a67\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\tThis classification can be further refined by grouping the current models into three groups:\n<ol>\n \t<li >The first group contains regression algorithms, decision trees, and traditional classification rules. They are <b>close to monotonic linear functions <\/b>and <b>commonly used in economics and sociology;<\/b><\/li>\n \t<li >The second group includes more advanced algorithms such as<b> graphical models<\/b><\/li>\n \t<li \">The third group consist of <\/span><b>advanced machine learning techniques such as SVM<\/b>, <b>ensemble learning, and deep learning methods.<\/b> They are only able to provide information on the importance of the variables for the explicability of the model.<\/li>\n<\/ol>\n&nbsp;\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-4fa81b2 elementor-widget elementor-widget-image\" data-id=\"4fa81b2\" data-element_type=\"widget\" data-widget_type=\"image.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t\t\t\t<figure class=\"wp-caption\">\n\t\t\t\t\t\t\t\t\t\t<img decoding=\"async\" width=\"800\" height=\"384\" src=\"https:\/\/datascientest.com\/wp-content\/uploads\/2020\/05\/Capture-d\u2019e\u0301cran-2020-05-20-a\u0300-13.02.44-1024x492.png\" class=\"attachment-large size-large wp-image-2815\" alt=\"Explicabilit\u00e9 vs interpr\u00e9tabilit\u00e9\" srcset=\"https:\/\/datascientest.com\/wp-content\/uploads\/2020\/05\/Capture-d\u2019e\u0301cran-2020-05-20-a\u0300-13.02.44-1024x492.png 1024w, https:\/\/datascientest.com\/wp-content\/uploads\/2020\/05\/Capture-d\u2019e\u0301cran-2020-05-20-a\u0300-13.02.44-300x144.png 300w, https:\/\/datascientest.com\/wp-content\/uploads\/2020\/05\/Capture-d\u2019e\u0301cran-2020-05-20-a\u0300-13.02.44-768x369.png 768w, https:\/\/datascientest.com\/wp-content\/uploads\/2020\/05\/Capture-d\u2019e\u0301cran-2020-05-20-a\u0300-13.02.44.png 1116w\" sizes=\"(max-width: 800px) 100vw, 800px\" \/>\t\t\t\t\t\t\t\t\t\t\t<figcaption class=\"widget-image-caption wp-caption-text\">Figure - Classification of Machine Learning Algorithms depending on their predictive power and Interpretability Adpated from [4]\n<\/figcaption>\n\t\t\t\t\t\t\t\t\t\t<\/figure>\n\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-474c7cf elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"474c7cf\" data-element_type=\"section\">\n\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"elementor-column elementor-col-100 elementor-top-column elementor-element elementor-element-e5840f3\" data-id=\"e5840f3\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-1963a52 elementor-widget elementor-widget-heading\" data-id=\"1963a52\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<h2 class=\"elementor-heading-title elementor-size-default\">Choosing the right model<\/h2>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-8747117 elementor-widget elementor-widget-text-editor\" data-id=\"8747117\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\tWhat is the right trade-off between model predictive power and interpretability?\nLet\u2019s review the main question you should explore:\n<ul>\n \t<li><b>Do you need a model with an intrinsic or post-hoc interpretability ? <br><\/b>Depending on the level of interpretability required, you might choose a model easy to explain (intrinsic interpretability) or a back-box model for the training and the use of the results such as the feature importance to explain (post-hoc interpretability).<\/span><\/li>\n \t<li><b >Do you need Model-Specific or Model-agnostic interpretability ? <\/b><br>There is a distinction to be made between the interpretability that can be specific to a model such as linear regression where weights can be interpreted and the interpretability that comes from general tools that applied after the model training (such as correlation, matrix confusion, or model-specific metrics)<\/span><\/li>\n \t<li><b >Do you need local or global interpretability ?<\/b><br>The global explicability requirement is intended to make the decision-making process transparent for all data, while the local explicability criterion is intended to provide explanations for a single decision in a restricted neighborhood of data.<\/span><\/li>\n<\/ul>\nTherefore, before starting a Data Science project, it is essential to first determine the degree of interpretability of the model we wish to achieve.\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-d32b3fd elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"d32b3fd\" data-element_type=\"section\">\n\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"elementor-column elementor-col-100 elementor-top-column elementor-element elementor-element-10f34ae\" data-id=\"10f34ae\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-80a1f4e elementor-widget elementor-widget-text-editor\" data-id=\"80a1f4e\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t<h4><span style=\"font-weight: normal;\">Do you want to learn more on Machine Learning algorithms?\u00a0<br \/>Feel free to <a href=\"contact@datascientest.com\">get in touch with us<\/a> and <a href=\"https:\/\/datascientest.com\/en\/home-page#formationsen\">learn more about our programmes<\/a>!\u00a0<\/span><\/h4>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-03e407f elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"03e407f\" data-element_type=\"section\">\n\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"elementor-column elementor-col-100 elementor-top-column elementor-element elementor-element-4ce1472\" data-id=\"4ce1472\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-c06a8c3 elementor-widget elementor-widget-heading\" data-id=\"c06a8c3\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<h2 class=\"elementor-heading-title elementor-size-default\">References\n\n\n<\/h2>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-39963a8 elementor-widget elementor-widget-text-editor\" data-id=\"39963a8\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t<span style=\"font-weight: 400;\">[1] C. Molnar, <\/span><i><span style=\"font-weight: 400;\">Interpretable Machine Learning<\/span><\/i>\n<br><a href=\"https:\/\/christophm.github.io\/interpretable-ml-book\/\"><span style=\"font-weight: 400;\">https:\/\/christophm.github.io\/interpretable-ml-book\/<\/span><\/a>\n<br><span style=\"font-weight: 400;\">[2] <\/span><i><span style=\"font-weight: 400;\">Probl\u00e9matiques juridiques et analyse automatique des donn\u00e9es,<\/span><\/i><span style=\"font-weight: 400;\"> Machine Learning and the Law<\/span>\n<br><a href=\"https:\/\/perso.math.univ-toulouse.fr\/mllaw\/home\/statisticien\/explicabilite-des-decisions-algorithmiques\/\"><span style=\"font-weight: 400;\">https:\/\/perso.math.univ-toulouse.fr\/mllaw\/home\/statisticien\/explicabilite-des-decisions-algorithmiques\/<\/span><\/a>\n\n<span style=\"font-weight: 400;\">[3] <\/span><span style=\"font-weight: 400;\">A. Verin\u00e9, S. Mir,<\/span><i><span style=\"font-weight: 400;\"> L\u2019interpr\u00e9tabilit\u00e9 du Machine learning: quels d\u00e9fis \u00e0 l\u2019\u00e8re des processus de la d\u00e9cisions automatis\u00e9s\u00a0?<\/span><\/i><span style=\"font-weight: 400;\"> Wavestone report<\/span>\n\n<a href=\"https:\/\/www.wavestone.com\/app\/uploads\/2019\/09\/Wavestone_Interpretabilite_Machine_learning.pdf\"><span style=\"font-weight: 400;\">https:\/\/www.wavestone.com\/app\/uploads\/2019\/09\/Wavestone_Interpretabilite_Machine_learning.pdf<\/span><\/a>\n\n<span style=\"font-weight: 400;\">[4] <\/span><span style=\"font-weight: 400;\">J. Cupe,<\/span><i><span style=\"font-weight: 400;\"> L\u2019interpr\u00e9tabilit\u00e9 de l\u2019IA \u2013 Le nouveau d\u00e9fi des data scientists, <\/span><\/i><span style=\"font-weight: 400;\">octobre 2018<\/span>\n<br><a href=\"https:\/\/www.actuia.com\/contribution\/jean-cupe\/linterpretabilite-de-lia-le-nouveau-defi-des-data-scientists\/\"><span style=\"font-weight: 400;\">https:\/\/www.actuia.com\/contribution\/jean-cupe\/linterpretabilite-de-lia-le-nouveau-defi-des-data-scientists\/<\/span><\/a>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t\t\t\t<\/div>\n\t\t","protected":false},"excerpt":{"rendered":"<p>Performance and interpretability: A necessary trade-off? Over the last decade, an increasing number of companies have embraced the digital transformation by incorporating Artificial Intelligence methods in the way they conceive their products and define their processes. Gathering, analyzing, and harnessing the data have been considered as essential drivers for business growth in an increasingly number [&hellip;]<\/p>\n","protected":false},"author":77,"featured_media":2824,"comment_status":"open","ping_status":"open","sticky":false,"template":"elementor_theme","format":"standard","meta":{"content-type":"","inline_featured_image":false},"categories":[2497],"tags":[162,163],"yoast_head":"<!-- This site is optimized with the Yoast SEO Premium plugin v21.7 (Yoast SEO v21.7) - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>Machine Learning : Performance and interpretability<\/title>\n<meta name=\"description\" content=\"Machine Learning: Understanding model interpretability to solve concrete problems is the first step towards successful Data Science projects.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/datascientest.com\/performance-and-interpretability-in-machine-learning\" \/>\n<meta property=\"og:locale\" content=\"fr_FR\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Machine Learning : Performance and interpretability\" \/>\n<meta property=\"og:url\" content=\"https:\/\/datascientest.com\/performance-and-interpretability-in-machine-learning\" \/>\n<meta property=\"og:site_name\" content=\"Formation Data Science | DataScientest.com\" \/>\n<meta property=\"article:published_time\" content=\"2020-05-25T08:47:32+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2023-11-09T14:14:33+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/datascientest.com\/wp-content\/uploads\/2020\/05\/Capture-d\u2019e\u0301cran-2020-05-20-a\u0300-13.24.49-e1589973950606.png\" \/>\n\t<meta property=\"og:image:width\" content=\"934\" \/>\n\t<meta property=\"og:image:height\" content=\"590\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/png\" \/>\n<meta name=\"author\" content=\"Raphael Kassel\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"\u00c9crit par\" \/>\n\t<meta name=\"twitter:data1\" content=\"Raphael Kassel\" \/>\n\t<meta name=\"twitter:label2\" content=\"Dur\u00e9e de lecture estim\u00e9e\" \/>\n\t<meta name=\"twitter:data2\" content=\"5 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\/\/datascientest.com\/performance-and-interpretability-in-machine-learning\",\"url\":\"https:\/\/datascientest.com\/performance-and-interpretability-in-machine-learning\",\"name\":\"Machine Learning : Performance and interpretability\",\"isPartOf\":{\"@id\":\"https:\/\/datascientest.com\/#website\"},\"datePublished\":\"2020-05-25T08:47:32+00:00\",\"dateModified\":\"2023-11-09T14:14:33+00:00\",\"author\":{\"@id\":\"https:\/\/datascientest.com\/#\/schema\/person\/6eca16f63daaafcbc96e15549b9d4ee1\"},\"description\":\"Machine Learning: Understanding model interpretability to solve concrete problems is the first step towards successful Data Science projects.\",\"inLanguage\":\"fr-FR\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/datascientest.com\/performance-and-interpretability-in-machine-learning\"]}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/datascientest.com\/#website\",\"url\":\"https:\/\/datascientest.com\/\",\"name\":\"Formation Data Science | DataScientest.com\",\"description\":\"Leader de la formation en Data Science\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/datascientest.com\/?s={search_term_string}\"},\"query-input\":\"required name=search_term_string\"}],\"inLanguage\":\"fr-FR\"},{\"@type\":\"Person\",\"@id\":\"https:\/\/datascientest.com\/#\/schema\/person\/6eca16f63daaafcbc96e15549b9d4ee1\",\"name\":\"Raphael Kassel\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"fr-FR\",\"@id\":\"https:\/\/datascientest.com\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/datascientest.com\/wp-content\/uploads\/2023\/11\/raph.jpeg\",\"contentUrl\":\"https:\/\/datascientest.com\/wp-content\/uploads\/2023\/11\/raph.jpeg\",\"caption\":\"Raphael Kassel\"},\"description\":\"Raphael Kassel est le Head of Data Programs chez DataScientest. Il supervise une \u00e9quipe Data, impl\u00e9mente de nouveaux parcours comme \\\"Chef de projet IA\\\" et \\\"Data Product Manager\\\", et contribue \u00e0 la modernisation des op\u00e9rations du groupe. En collaboration \u00e9troite avec l'\u00e9quipe de d\u00e9veloppement, il adapte l'accompagnement acad\u00e9mique aux \u00e9volutions de l'\u00e9quipe data et \u00e0 l'internationalisation.\",\"sameAs\":[\"https:\/\/www.linkedin.com\/in\/raphaelkassel\/\",\"Data Science\"],\"url\":\"https:\/\/datascientest.com\/author\/raphaelkassel\"}]}<\/script>\n<!-- \/ Yoast SEO Premium plugin. -->","yoast_head_json":{"title":"Machine Learning : Performance and interpretability","description":"Machine Learning: Understanding model interpretability to solve concrete problems is the first step towards successful Data Science projects.","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:\/\/datascientest.com\/performance-and-interpretability-in-machine-learning","og_locale":"fr_FR","og_type":"article","og_title":"Machine Learning : Performance and interpretability","og_url":"https:\/\/datascientest.com\/performance-and-interpretability-in-machine-learning","og_site_name":"Formation Data Science | DataScientest.com","article_published_time":"2020-05-25T08:47:32+00:00","article_modified_time":"2023-11-09T14:14:33+00:00","og_image":[{"width":934,"height":590,"url":"https:\/\/datascientest.com\/wp-content\/uploads\/2020\/05\/Capture-d\u2019e\u0301cran-2020-05-20-a\u0300-13.24.49-e1589973950606.png","type":"image\/png"}],"author":"Raphael Kassel","twitter_card":"summary_large_image","twitter_misc":{"\u00c9crit par":"Raphael Kassel","Dur\u00e9e de lecture estim\u00e9e":"5 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/datascientest.com\/performance-and-interpretability-in-machine-learning","url":"https:\/\/datascientest.com\/performance-and-interpretability-in-machine-learning","name":"Machine Learning : Performance and interpretability","isPartOf":{"@id":"https:\/\/datascientest.com\/#website"},"datePublished":"2020-05-25T08:47:32+00:00","dateModified":"2023-11-09T14:14:33+00:00","author":{"@id":"https:\/\/datascientest.com\/#\/schema\/person\/6eca16f63daaafcbc96e15549b9d4ee1"},"description":"Machine Learning: Understanding model interpretability to solve concrete problems is the first step towards successful Data Science projects.","inLanguage":"fr-FR","potentialAction":[{"@type":"ReadAction","target":["https:\/\/datascientest.com\/performance-and-interpretability-in-machine-learning"]}]},{"@type":"WebSite","@id":"https:\/\/datascientest.com\/#website","url":"https:\/\/datascientest.com\/","name":"Formation Data Science | DataScientest.com","description":"Leader de la formation en Data Science","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/datascientest.com\/?s={search_term_string}"},"query-input":"required name=search_term_string"}],"inLanguage":"fr-FR"},{"@type":"Person","@id":"https:\/\/datascientest.com\/#\/schema\/person\/6eca16f63daaafcbc96e15549b9d4ee1","name":"Raphael Kassel","image":{"@type":"ImageObject","inLanguage":"fr-FR","@id":"https:\/\/datascientest.com\/#\/schema\/person\/image\/","url":"https:\/\/datascientest.com\/wp-content\/uploads\/2023\/11\/raph.jpeg","contentUrl":"https:\/\/datascientest.com\/wp-content\/uploads\/2023\/11\/raph.jpeg","caption":"Raphael Kassel"},"description":"Raphael Kassel est le Head of Data Programs chez DataScientest. Il supervise une \u00e9quipe Data, impl\u00e9mente de nouveaux parcours comme \"Chef de projet IA\" et \"Data Product Manager\", et contribue \u00e0 la modernisation des op\u00e9rations du groupe. En collaboration \u00e9troite avec l'\u00e9quipe de d\u00e9veloppement, il adapte l'accompagnement acad\u00e9mique aux \u00e9volutions de l'\u00e9quipe data et \u00e0 l'internationalisation.","sameAs":["https:\/\/www.linkedin.com\/in\/raphaelkassel\/","Data Science"],"url":"https:\/\/datascientest.com\/author\/raphaelkassel"}]}},"_links":{"self":[{"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/posts\/2934"}],"collection":[{"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/users\/77"}],"replies":[{"embeddable":true,"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/comments?post=2934"}],"version-history":[{"count":0,"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/posts\/2934\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/media\/2824"}],"wp:attachment":[{"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/media?parent=2934"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/categories?post=2934"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/tags?post=2934"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}