{"id":3435,"date":"2021-06-30T09:43:59","date_gmt":"2021-06-30T09:43:59","guid":{"rendered":"https:\/\/staging.novelis.io\/?post_type=scientific-pub&#038;p=3435"},"modified":"2022-02-25T10:48:05","modified_gmt":"2022-02-25T10:48:05","slug":"generation-de-requetes-sql-a-partir-du-nl-un-modele-seq2seq-transformers-architecture","status":"publish","type":"scientific-pub","link":"https:\/\/novelis.io\/fr\/scientific-pub\/generation-de-requetes-sql-a-partir-du-nl-un-modele-seq2seq-transformers-architecture\/","title":{"rendered":"G\u00e9n\u00e9ration de requ\u00eates SQL \u00e0 partir du NL : un mod\u00e8le Seq2Seq \u2013 Transformers Architecture"},"content":{"rendered":"\n<p>Les experts techniques de Novelis ont une fois de plus atteint un nouvel \u00e9tat de l\u2019art scientifique. D\u00e9couvrez notre \u00e9tude&nbsp;<strong><a href=\"https:\/\/doi.org\/10.3844\/jcssp.2021.480.489\" target=\"_blank\" rel=\"noopener\">SQL Generation from Natural Language: A Sequence-to-Sequence Model Powered by the Transformers Architecture and Association Rules<\/a><\/strong>, publi\u00e9e dans le<em>&nbsp;Journal of Computer Science<\/em>. Cette \u00e9tude est en anglais.<\/p>\n\n\n\n<p>Merci \u00e0 l\u2019<a href=\"https:\/\/novelis.io\/fr\/laboratoire-rd\/\">\u00e9quipe de recherche de Novelis<\/a> pour son savoir-faire et son expertise.<\/p>\n\n\n\n<div style=\"height:31px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n\n\n\n<h2 class=\"wp-block-heading\" id=\"resume\">R\u00e9sum\u00e9<\/h2>\n\n\n\n<p>L\u2019utilisation du langage naturel (NL) pour interagir avec les bases de donn\u00e9es relationnelles permet aux utilisateurs de tous niveaux d\u2019interroger et d\u2019analyser facilement de grandes quantit\u00e9s de donn\u00e9es. Cela n\u00e9cessite un syst\u00e8me qui comprend les questions des utilisateurs et les traduit automatiquement en langages de requ\u00eates structur\u00e9s (tels que SQL). Le syst\u00e8me Text-to-SQL le plus performant utilise l\u2019apprentissage supervis\u00e9 (g\u00e9n\u00e9ralement exprim\u00e9 sous la forme d\u2019un probl\u00e8me de classification) et traite cette t\u00e2che comme un probl\u00e8me de remplissage de cases bas\u00e9 sur des croquis, ou convertit d\u2019abord le probl\u00e8me en une forme logique interm\u00e9diaire (ILF), puis le convertit en la requ\u00eate SQL correspondante. Cependant, la mod\u00e9lisation non supervis\u00e9e qui traduit directement le probl\u00e8me en requ\u00eates SQL s\u2019est av\u00e9r\u00e9e plus difficile. Dans ce sens, nous proposons une m\u00e9thode pour convertir directement les questions NL en requ\u00eates SQL.<\/p>\n\n\n\n<p>Dans cette recherche, nous proposons un mod\u00e8le d\u2019analyse syntaxique de s\u00e9quence \u00e0 s\u00e9quence (Seq2Seq) pour les t\u00e2ches NL vers SQL, soutenu par une architecture de convertisseur qui explore deux mod\u00e8les de langage (LM) : le convertisseur de transfert de texte \u00e0 texte (T5). Et le convertisseur texte-texte pr\u00e9-entra\u00een\u00e9 multi-langue (mT5). En outre, nous utilisons des algorithmes d\u2019apprentissage bas\u00e9s sur la transformation pour mettre \u00e0 jour les pr\u00e9dictions d\u2019agr\u00e9gation bas\u00e9es sur les r\u00e8gles d\u2019association. Le mod\u00e8le r\u00e9sultant met en \u0153uvre une nouvelle technologie de pointe sur le jeu de donn\u00e9es WikiSQL pour la g\u00e9n\u00e9ration de SQL faiblement supervis\u00e9e.<\/p>\n\n\n\n<div style=\"height:29px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n\n\n\n<h2 class=\"wp-block-heading\" id=\"a-propos-de-l-etude\">A propos de l\u2019\u00e9tude<\/h2>\n\n\n\n<p>\u201cIn this study, we treat the Text-to-SQL task with WikiSQL1 (Zhong et al., 2017). This DataSet is the first large-scale dataset for Text-to-SQL, with about 80 K human-annotated pairs of Natural Language question and SQL query. WikiSQL is very challenging because tables and questions are very diverse. This DataSet contains about 24K different tables.<\/p>\n\n\n\n<p>There are two leaderboards for the WikiSQL challenge: Weakly supervised (without using logical form during training) and supervised (with logical form during training). On the supervised challenge, there are two results: Those with Execution Guided (EG) inference and those without EG inference.\u201d<\/p>\n\n\n\n<div style=\"height:31px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n\n\n\n<h3 class=\"wp-block-heading\" id=\"lire-l-article-complet\"><br><a href=\"https:\/\/thescipub.com\/abstract\/10.3844\/jcssp.2021.480.489\" target=\"_blank\" rel=\"noopener\">Lire l\u2019article complet<\/a><\/h3>\n\n\n\n<p><strong>Journal of Computer Science<\/strong>&nbsp;\u2013 Volume 17 n\u00b0 5, 2021, 480-489 (10 pages)<\/p>\n\n\n\n<p><strong>Le Journal of Computer Science a pour objectif de publier des articles de recherche sur les bases th\u00e9oriques de l\u2019information et de l\u2019informatique, ainsi que sur les technologies pratiques de mise en \u0153uvre et d\u2019application dans les syst\u00e8mes informatiques.<\/strong><\/p>\n","protected":false},"featured_media":2292,"template":"","categories":[24],"custom_tag":[165,167],"class_list":["post-3435","scientific-pub","type-scientific-pub","status-publish","has-post-thumbnail","hentry","category-publication-scientifique","custom_tag-nlp-fr","custom_tag-sql-fr"],"acf":{"summary":"","filter_opacity":"70","subtitle":"","reading_time":"","authors":"","document_to_download":{"upload_a_file":false,"download_without_form":false,"file":false,"url":""},"externel_link":""},"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v25.6 - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>Requ\u00eates SQL \u00e0 partir du NL : Seq2Seq \u2013 Transformers Architecture<\/title>\n<meta name=\"description\" content=\"D\u00e9couvrez notre \u00e9tude\u00a0SQL Generation from NL: A Sequence-to-Sequence Model Powered by the Transformers Architecture and Association Rules.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/novelis.io\/fr\/scientific-pub\/generation-de-requetes-sql-a-partir-du-nl-un-modele-seq2seq-transformers-architecture\/\" \/>\n<meta property=\"og:locale\" content=\"fr_FR\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Requ\u00eates SQL \u00e0 partir du NL : Seq2Seq \u2013 Transformers Architecture\" \/>\n<meta property=\"og:description\" content=\"D\u00e9couvrez notre \u00e9tude\u00a0SQL Generation from NL: A Sequence-to-Sequence Model Powered by the Transformers Architecture and Association Rules.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/novelis.io\/fr\/scientific-pub\/generation-de-requetes-sql-a-partir-du-nl-un-modele-seq2seq-transformers-architecture\/\" \/>\n<meta property=\"og:site_name\" content=\"Novelis innovation\" \/>\n<meta property=\"article:publisher\" content=\"https:\/\/www.facebook.com\/novelis.io\" \/>\n<meta property=\"article:modified_time\" content=\"2022-02-25T10:48:05+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/novelis.io\/wp-content\/uploads\/2022\/02\/images-9.png\" \/>\n\t<meta property=\"og:image:width\" content=\"1200\" \/>\n\t<meta property=\"og:image:height\" content=\"628\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/png\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:site\" content=\"@novelis_io\" \/>\n<meta name=\"twitter:label1\" content=\"Dur\u00e9e de lecture estim\u00e9e\" \/>\n\t<meta name=\"twitter:data1\" content=\"2 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\/\/novelis.io\/fr\/scientific-pub\/generation-de-requetes-sql-a-partir-du-nl-un-modele-seq2seq-transformers-architecture\/\",\"url\":\"https:\/\/novelis.io\/fr\/scientific-pub\/generation-de-requetes-sql-a-partir-du-nl-un-modele-seq2seq-transformers-architecture\/\",\"name\":\"Requ\u00eates SQL \u00e0 partir du NL : Seq2Seq \u2013 Transformers Architecture\",\"isPartOf\":{\"@id\":\"https:\/\/novelis.io\/fr\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\/\/novelis.io\/fr\/scientific-pub\/generation-de-requetes-sql-a-partir-du-nl-un-modele-seq2seq-transformers-architecture\/#primaryimage\"},\"image\":{\"@id\":\"https:\/\/novelis.io\/fr\/scientific-pub\/generation-de-requetes-sql-a-partir-du-nl-un-modele-seq2seq-transformers-architecture\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/novelis.io\/wp-content\/uploads\/2022\/02\/images-9.png\",\"datePublished\":\"2021-06-30T09:43:59+00:00\",\"dateModified\":\"2022-02-25T10:48:05+00:00\",\"description\":\"D\u00e9couvrez notre \u00e9tude\u00a0SQL Generation from NL: A Sequence-to-Sequence Model Powered by the Transformers Architecture and Association Rules.\",\"breadcrumb\":{\"@id\":\"https:\/\/novelis.io\/fr\/scientific-pub\/generation-de-requetes-sql-a-partir-du-nl-un-modele-seq2seq-transformers-architecture\/#breadcrumb\"},\"inLanguage\":\"fr-FR\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/novelis.io\/fr\/scientific-pub\/generation-de-requetes-sql-a-partir-du-nl-un-modele-seq2seq-transformers-architecture\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"fr-FR\",\"@id\":\"https:\/\/novelis.io\/fr\/scientific-pub\/generation-de-requetes-sql-a-partir-du-nl-un-modele-seq2seq-transformers-architecture\/#primaryimage\",\"url\":\"https:\/\/novelis.io\/wp-content\/uploads\/2022\/02\/images-9.png\",\"contentUrl\":\"https:\/\/novelis.io\/wp-content\/uploads\/2022\/02\/images-9.png\",\"width\":1200,\"height\":628},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/novelis.io\/fr\/scientific-pub\/generation-de-requetes-sql-a-partir-du-nl-un-modele-seq2seq-transformers-architecture\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Accueil\",\"item\":\"https:\/\/novelis.io\/fr\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"G\u00e9n\u00e9ration de requ\u00eates SQL \u00e0 partir du NL : un mod\u00e8le Seq2Seq \u2013 Transformers Architecture\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/novelis.io\/fr\/#website\",\"url\":\"https:\/\/novelis.io\/fr\/\",\"name\":\"Novelis innovation\",\"description\":\"Novelis innovation\",\"publisher\":{\"@id\":\"https:\/\/novelis.io\/fr\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/novelis.io\/fr\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"fr-FR\"},{\"@type\":\"Organization\",\"@id\":\"https:\/\/novelis.io\/fr\/#organization\",\"name\":\"Novelis innovation\",\"url\":\"https:\/\/novelis.io\/fr\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"fr-FR\",\"@id\":\"https:\/\/novelis.io\/fr\/#\/schema\/logo\/image\/\",\"url\":\"https:\/\/novelis.io\/wp-content\/uploads\/2021\/12\/logo-1.png\",\"contentUrl\":\"https:\/\/novelis.io\/wp-content\/uploads\/2021\/12\/logo-1.png\",\"width\":479,\"height\":98,\"caption\":\"Novelis innovation\"},\"image\":{\"@id\":\"https:\/\/novelis.io\/fr\/#\/schema\/logo\/image\/\"},\"sameAs\":[\"https:\/\/www.facebook.com\/novelis.io\",\"https:\/\/x.com\/novelis_io\",\"https:\/\/www.linkedin.com\/company\/novelis-consulting\/\",\"https:\/\/www.youtube.com\/channel\/UCJ5eJR22n2GtfKaTWueWRPQ\"]}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Requ\u00eates SQL \u00e0 partir du NL : Seq2Seq \u2013 Transformers Architecture","description":"D\u00e9couvrez notre \u00e9tude\u00a0SQL Generation from NL: A Sequence-to-Sequence Model Powered by the Transformers Architecture and Association Rules.","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:\/\/novelis.io\/fr\/scientific-pub\/generation-de-requetes-sql-a-partir-du-nl-un-modele-seq2seq-transformers-architecture\/","og_locale":"fr_FR","og_type":"article","og_title":"Requ\u00eates SQL \u00e0 partir du NL : Seq2Seq \u2013 Transformers Architecture","og_description":"D\u00e9couvrez notre \u00e9tude\u00a0SQL Generation from NL: A Sequence-to-Sequence Model Powered by the Transformers Architecture and Association Rules.","og_url":"https:\/\/novelis.io\/fr\/scientific-pub\/generation-de-requetes-sql-a-partir-du-nl-un-modele-seq2seq-transformers-architecture\/","og_site_name":"Novelis innovation","article_publisher":"https:\/\/www.facebook.com\/novelis.io","article_modified_time":"2022-02-25T10:48:05+00:00","og_image":[{"width":1200,"height":628,"url":"https:\/\/novelis.io\/wp-content\/uploads\/2022\/02\/images-9.png","type":"image\/png"}],"twitter_card":"summary_large_image","twitter_site":"@novelis_io","twitter_misc":{"Dur\u00e9e de lecture estim\u00e9e":"2 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/novelis.io\/fr\/scientific-pub\/generation-de-requetes-sql-a-partir-du-nl-un-modele-seq2seq-transformers-architecture\/","url":"https:\/\/novelis.io\/fr\/scientific-pub\/generation-de-requetes-sql-a-partir-du-nl-un-modele-seq2seq-transformers-architecture\/","name":"Requ\u00eates SQL \u00e0 partir du NL : Seq2Seq \u2013 Transformers Architecture","isPartOf":{"@id":"https:\/\/novelis.io\/fr\/#website"},"primaryImageOfPage":{"@id":"https:\/\/novelis.io\/fr\/scientific-pub\/generation-de-requetes-sql-a-partir-du-nl-un-modele-seq2seq-transformers-architecture\/#primaryimage"},"image":{"@id":"https:\/\/novelis.io\/fr\/scientific-pub\/generation-de-requetes-sql-a-partir-du-nl-un-modele-seq2seq-transformers-architecture\/#primaryimage"},"thumbnailUrl":"https:\/\/novelis.io\/wp-content\/uploads\/2022\/02\/images-9.png","datePublished":"2021-06-30T09:43:59+00:00","dateModified":"2022-02-25T10:48:05+00:00","description":"D\u00e9couvrez notre \u00e9tude\u00a0SQL Generation from NL: A Sequence-to-Sequence Model Powered by the Transformers Architecture and Association Rules.","breadcrumb":{"@id":"https:\/\/novelis.io\/fr\/scientific-pub\/generation-de-requetes-sql-a-partir-du-nl-un-modele-seq2seq-transformers-architecture\/#breadcrumb"},"inLanguage":"fr-FR","potentialAction":[{"@type":"ReadAction","target":["https:\/\/novelis.io\/fr\/scientific-pub\/generation-de-requetes-sql-a-partir-du-nl-un-modele-seq2seq-transformers-architecture\/"]}]},{"@type":"ImageObject","inLanguage":"fr-FR","@id":"https:\/\/novelis.io\/fr\/scientific-pub\/generation-de-requetes-sql-a-partir-du-nl-un-modele-seq2seq-transformers-architecture\/#primaryimage","url":"https:\/\/novelis.io\/wp-content\/uploads\/2022\/02\/images-9.png","contentUrl":"https:\/\/novelis.io\/wp-content\/uploads\/2022\/02\/images-9.png","width":1200,"height":628},{"@type":"BreadcrumbList","@id":"https:\/\/novelis.io\/fr\/scientific-pub\/generation-de-requetes-sql-a-partir-du-nl-un-modele-seq2seq-transformers-architecture\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Accueil","item":"https:\/\/novelis.io\/fr\/"},{"@type":"ListItem","position":2,"name":"G\u00e9n\u00e9ration de requ\u00eates SQL \u00e0 partir du NL : un mod\u00e8le Seq2Seq \u2013 Transformers Architecture"}]},{"@type":"WebSite","@id":"https:\/\/novelis.io\/fr\/#website","url":"https:\/\/novelis.io\/fr\/","name":"Novelis innovation","description":"Novelis innovation","publisher":{"@id":"https:\/\/novelis.io\/fr\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/novelis.io\/fr\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"fr-FR"},{"@type":"Organization","@id":"https:\/\/novelis.io\/fr\/#organization","name":"Novelis innovation","url":"https:\/\/novelis.io\/fr\/","logo":{"@type":"ImageObject","inLanguage":"fr-FR","@id":"https:\/\/novelis.io\/fr\/#\/schema\/logo\/image\/","url":"https:\/\/novelis.io\/wp-content\/uploads\/2021\/12\/logo-1.png","contentUrl":"https:\/\/novelis.io\/wp-content\/uploads\/2021\/12\/logo-1.png","width":479,"height":98,"caption":"Novelis innovation"},"image":{"@id":"https:\/\/novelis.io\/fr\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/www.facebook.com\/novelis.io","https:\/\/x.com\/novelis_io","https:\/\/www.linkedin.com\/company\/novelis-consulting\/","https:\/\/www.youtube.com\/channel\/UCJ5eJR22n2GtfKaTWueWRPQ"]}]}},"_links":{"self":[{"href":"https:\/\/novelis.io\/fr\/wp-json\/wp\/v2\/scientific-pub\/3435"}],"collection":[{"href":"https:\/\/novelis.io\/fr\/wp-json\/wp\/v2\/scientific-pub"}],"about":[{"href":"https:\/\/novelis.io\/fr\/wp-json\/wp\/v2\/types\/scientific-pub"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/novelis.io\/fr\/wp-json\/wp\/v2\/media\/2292"}],"wp:attachment":[{"href":"https:\/\/novelis.io\/fr\/wp-json\/wp\/v2\/media?parent=3435"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/novelis.io\/fr\/wp-json\/wp\/v2\/categories?post=3435"},{"taxonomy":"custom_tag","embeddable":true,"href":"https:\/\/novelis.io\/fr\/wp-json\/wp\/v2\/custom_tag?post=3435"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}