<?xml version='1.0' encoding='UTF-8'?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:ccmbenchmark="https://www.ccmbenchmark.com/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:media="http://search.yahoo.com/mrss/"><channel><title>Paul Sina&amp;iuml; : Dernier contenus</title><link>/account/paul-sinai-27015</link><atom:link href="/account/paul-sinai-27015" rel="self" type="application/rss+xml" /><language>fr</language><pubDate>Mon, 27 Apr 2026 23:45:12 +0200</pubDate><lastBuildDate>Mon, 27 Apr 2026 23:45:12 +0200</lastBuildDate><item><title>Projets data : quatre écueils qui limitent le ROI de la data science</title><link>https://www.journaldunet.com/intelligence-artificielle/1208941-projets-data-quatre-ecueils-qui-limitent-le-roi-de-la-data-science/</link><description><![CDATA[<a href="https://www.journaldunet.com/intelligence-artificielle/1208941-projets-data-quatre-ecueils-qui-limitent-le-roi-de-la-data-science/"><img src="https://img-0.journaldunet.com/qG2FKaElXlTwrGAmPtQZJBqSOUY=/100x/smart/20fa970ee2ad40eaa6a9624d248f2fd5/user-jdn/10023888-anonyme-anonyme.jpg" align="left" hspace="5" vspace="0"></a>C’est un fait, un trop grand nombre de projets data piétinent au point de ne pas dépasser le stade du proof-of-concept. En cause, quatre dérives récurrentes qu’il est pourtant aujourd'hui possible d’éviter.]]></description><pubDate>Tue, 17 Apr 2018 12:05:51 +0200</pubDate><guid>https://www.journaldunet.com/intelligence-artificielle/1208941-projets-data-quatre-ecueils-qui-limitent-le-roi-de-la-data-science/</guid><dc:creator><![CDATA[Paul Sina&iuml;]]></dc:creator><ccmbenchmark:content_author><![CDATA[Paul Sina&iuml;]]></ccmbenchmark:content_author><ccmbenchmark:content_author_company><![CDATA[ForePaaS]]></ccmbenchmark:content_author_company><ccmbenchmark:content_author_illustration>1</ccmbenchmark:content_author_illustration></item></channel></rss>