{"id":16569,"date":"2024-09-20T12:19:15","date_gmt":"2024-09-20T10:19:15","guid":{"rendered":"https:\/\/is.ijs.si\/?p=16569"},"modified":"2025-03-26T13:15:41","modified_gmt":"2025-03-26T12:15:41","slug":"integrating-knowledge-graphs-and-large-language-models-for-querying-in-an-industrial-environment","status":"publish","type":"post","link":"https:\/\/is.ijs.si\/?p=16569","title":{"rendered":"Integrating Knowledge Graphs and Large Language Models for Querying in an Industrial Environment"},"content":{"rendered":"\n<p>Domen Ho\u010devar and Klemen Kenda<\/p>\n<p>Abstract<br \/>Knowledge graphs have traditionally required the use of specific<br \/>query languages, such as SPARQL, to retrieve relevant data. In<br \/>this paper, we present a system capable of performing natural<br \/>language queries on knowledge graphs by leveraging retrieval-<br \/>augmented generation (RAG) and large language models (LLMs).<br \/>Our system can ingest large knowledge graphs and answer queries<br \/>using two approaches: first, by utilizing LLMs to extract informa-<br \/>tion directly from subgraphs; and second, by generating SPARQL<br \/>queries with LLMs and using the results to inform further infer-<br \/>ence, such as counting the number of items.<\/p>\n<p>\u00a0<\/p>\n\n\n\n<div data-wp-interactive=\"core\/file\" class=\"wp-block-file\"><object data-wp-bind--hidden=\"!state.hasPdfPreview\" hidden class=\"wp-block-file__embed\" data=\"https:\/\/is.ijs.si\/wp-content\/uploads\/2024\/10\/IS2024_-_SIKDD_2024_paper_5-1.pdf\" type=\"application\/pdf\" style=\"width:100%;height:600px\" aria-label=\"Embed of IS2024_-_SIKDD_2024_paper_5-1.\"><\/object><a id=\"wp-block-file--media-55f1ec2b-cf79-470f-a9dd-1616e8ebb879\" href=\"https:\/\/is.ijs.si\/wp-content\/uploads\/2024\/10\/IS2024_-_SIKDD_2024_paper_5-1.pdf\">IS2024_-_SIKDD_2024_paper_5-1<\/a><a href=\"https:\/\/is.ijs.si\/wp-content\/uploads\/2024\/10\/IS2024_-_SIKDD_2024_paper_5-1.pdf\" class=\"wp-block-file__button wp-element-button\" download aria-describedby=\"wp-block-file--media-55f1ec2b-cf79-470f-a9dd-1616e8ebb879\">Download<\/a><\/div>\n","protected":false},"excerpt":{"rendered":"","protected":false},"author":29,"featured_media":24966,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[109,102],"tags":[],"class_list":["post-16569","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-doi-sikdd-2024","category-papers"],"_links":{"self":[{"href":"https:\/\/is.ijs.si\/index.php?rest_route=\/wp\/v2\/posts\/16569","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/is.ijs.si\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/is.ijs.si\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/is.ijs.si\/index.php?rest_route=\/wp\/v2\/users\/29"}],"replies":[{"embeddable":true,"href":"https:\/\/is.ijs.si\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=16569"}],"version-history":[{"count":2,"href":"https:\/\/is.ijs.si\/index.php?rest_route=\/wp\/v2\/posts\/16569\/revisions"}],"predecessor-version":[{"id":16930,"href":"https:\/\/is.ijs.si\/index.php?rest_route=\/wp\/v2\/posts\/16569\/revisions\/16930"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/is.ijs.si\/index.php?rest_route=\/wp\/v2\/media\/24966"}],"wp:attachment":[{"href":"https:\/\/is.ijs.si\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=16569"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/is.ijs.si\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=16569"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/is.ijs.si\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=16569"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}