From Literature to Knowledge Graphs: Automated Extraction and Retrieval of Life Cycle Assessment Data with Large Language Models

From Literature to Knowledge Graphs: Automated Extraction and Retrieval of Life Cycle Assessment Data with Large Language Models | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article From Literature to Knowledge Graphs: Automated Extraction and Retrieval of Life Cycle Assessment Data with Large Language Models Zirui Tang, Max Dreger, Peijin Jiang, Kourosh Malek, Qingshi Tu This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8633241/v1 This work is licensed under a CC BY 4.0 License Status: Under Revision Version 1 posted 7 You are reading this latest preprint version Abstract The large and steadily growing volume of scientific literature poses challenges for accessing and utilizing data due to its unstructured nature. Life cycle assessment (LCA), in particular, relies on high-quality life cycle inventory (LCI) data to quantify the environmental impacts of products or services across their life cycle, yet manual extraction of such data is time-consuming and labor-intensive. This study presents a framework that integrates large language models (LLMs) with knowledge graph (KG) to extract and manage LCA data from published studies. A retrieval-augmented generation (RAG) pipeline is developed to mine three core data types from full-text articles: LCI data, life cycle impact assessment (LCIA) results, and LCA modeling assumptions. The extracted information is normalized and mapped to an ontology-driven, LCA-oriented knowledge graph (LCAKG) implemented in Neo4j. To support user interaction, an LLM-based question-answering system translates natural language queries into executable graph queries, allowing users to retrieve rich information without prior knowledge of KG schemas. The framework is evaluated using a case study of LCA studies on chemical production. The results demonstrate high semantic accuracy in data extraction, with F1-scores ranging from 73.54% to 93.34%. Query performance is significantly improved by combining similarity search with text-to-cypher reasoning, increasing the F1-score from 56.98% (baseline) to 75.18%. The proposed framework enhances the accessibility and interoperability of LCA domain data and provides a scalable foundation for large-scale knowledge synthesis to support sustainability research. Life cycle assessment Large language models Machine Learning Data Extraction Knowledge graph Full Text Additional Declarations No competing interests reported. Supplementary Files SI12.30cleanversion.docx SupplementarydataS1.xlsx SupplementarydataS2.xlsx SupplementarydataS3.xlsx Cite Share Download PDF Status: Under Revision Version 1 posted Editorial decision: Revision requested 09 Apr, 2026 Reviews received at journal 09 Mar, 2026 Reviewers agreed at journal 09 Feb, 2026 Reviewers invited by journal 03 Feb, 2026 Editor assigned by journal 21 Jan, 2026 Submission checks completed at journal 21 Jan, 2026 First submitted to journal 18 Jan, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8633241","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":578254625,"identity":"0efb2d15-30ca-48f4-9b07-1ffe5aa87cb5","order_by":0,"name":"Zirui Tang","email":"","orcid":"","institution":"University of British Columbia","correspondingAuthor":false,"prefix":"","firstName":"Zirui","middleName":"","lastName":"Tang","suffix":""},{"id":578254626,"identity":"5978c16b-86f2-48db-9514-1771765167df","order_by":1,"name":"Max Dreger","email":"","orcid":"","institution":"Forschungszentrum Juelich GmbH","correspondingAuthor":false,"prefix":"","firstName":"Max","middleName":"","lastName":"Dreger","suffix":""},{"id":578254627,"identity":"14469668-7de6-443b-81db-1cb05530129e","order_by":2,"name":"Peijin Jiang","email":"","orcid":"","institution":"University of British Columbia","correspondingAuthor":false,"prefix":"","firstName":"Peijin","middleName":"","lastName":"Jiang","suffix":""},{"id":578254628,"identity":"e7669256-7a1f-447c-83ae-03339eda7765","order_by":3,"name":"Kourosh Malek","email":"","orcid":"","institution":"Forschungszentrum Juelich GmbH","correspondingAuthor":false,"prefix":"","firstName":"Kourosh","middleName":"","lastName":"Malek","suffix":""},{"id":578254629,"identity":"72b78b59-1103-4254-9fff-1f6775d6daeb","order_by":4,"name":"Qingshi Tu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAxUlEQVRIiWNgGAWjYDACZiBmbJAAsQ4wk6aFh4EtgUgtDGAtDEAtPAbEaeE7znv45c8dFgz2EjnfpAtqGOT5GwhokTzMl2bNewboMIncbdIzjjEYzjhAQIvBYR4zY8Y2qBbeBoYEBmK0GP4Ea8l5BtYiT4QW4we8EC1sYC0GhLRIAm1hBmrh4TnzzNh6xjEJw42EtPCdP2P88WdbnRx7e/LD2wU1NvJyhLQAPcsGinoeKFeCkHqwFuYPRCgbBaNgFIyCkQwA8ps2MKgmQ98AAAAASUVORK5CYII=","orcid":"","institution":"University of British Columbia","correspondingAuthor":true,"prefix":"","firstName":"Qingshi","middleName":"","lastName":"Tu","suffix":""}],"badges":[],"createdAt":"2026-01-18 19:08:25","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8633241/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8633241/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":101751160,"identity":"f561dff2-8bb8-41b8-a847-580417e83712","added_by":"auto","created_at":"2026-02-03 10:17:05","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1138119,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript12.30cleanversion.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8633241/v1_covered_7665dfc7-2a40-47db-8807-f5ba8270099d.pdf"},{"id":101297600,"identity":"695daca0-e9a4-48bd-81b0-2cd2636ddff8","added_by":"auto","created_at":"2026-01-28 09:28:12","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":6227169,"visible":true,"origin":"","legend":"","description":"","filename":"SI12.30cleanversion.docx","url":"https://assets-eu.researchsquare.com/files/rs-8633241/v1/f254e2b9c10efa52387eff73.docx"},{"id":101249420,"identity":"75201ec4-aaa7-4165-95f1-c32eb10d01b5","added_by":"auto","created_at":"2026-01-27 17:19:35","extension":"xlsx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":754921,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementarydataS1.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-8633241/v1/128e7707af6d4985cd7d4536.xlsx"},{"id":101297388,"identity":"991ec7e7-5536-40f9-8ed2-b069ddc96f2d","added_by":"auto","created_at":"2026-01-28 09:26:59","extension":"xlsx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":463627,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementarydataS2.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-8633241/v1/555dda60845fc01e349dfec5.xlsx"},{"id":101297682,"identity":"d1c6975e-93de-4ccd-a7a3-8244e05c0746","added_by":"auto","created_at":"2026-01-28 09:28:33","extension":"xlsx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":154917,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementarydataS3.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-8633241/v1/70f7691a74a68a59b855c296.xlsx"}],"financialInterests":"No competing interests reported.","formattedTitle":"From Literature to Knowledge Graphs: Automated Extraction and Retrieval of Life Cycle Assessment Data with Large Language Models","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"journal-of-industrial-ecology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"44498","submissionUrl":"https://submission.springernature.com/new-submission/44498/3","title":"Journal of Industrial Ecology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Life cycle assessment, Large language models, Machine Learning, Data Extraction, Knowledge graph","lastPublishedDoi":"10.21203/rs.3.rs-8633241/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8633241/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe large and steadily growing volume of scientific literature poses challenges for accessing and utilizing data due to its unstructured nature. Life cycle assessment (LCA), in particular, relies on high-quality life cycle inventory (LCI) data to quantify the environmental impacts of products or services across their life cycle, yet manual extraction of such data is time-consuming and labor-intensive. This study presents a framework that integrates large language models (LLMs) with knowledge graph (KG) to extract and manage LCA data from published studies. A retrieval-augmented generation (RAG) pipeline is developed to mine three core data types from full-text articles: LCI data, life cycle impact assessment (LCIA) results, and LCA modeling assumptions. The extracted information is normalized and mapped to an ontology-driven, LCA-oriented knowledge graph (LCAKG) implemented in Neo4j. To support user interaction, an LLM-based question-answering system translates natural language queries into executable graph queries, allowing users to retrieve rich information without prior knowledge of KG schemas. The framework is evaluated using a case study of LCA studies on chemical production. The results demonstrate high semantic accuracy in data extraction, with F1-scores ranging from 73.54% to 93.34%. Query performance is significantly improved by combining similarity search with text-to-cypher reasoning, increasing the F1-score from 56.98% (baseline) to 75.18%. The proposed framework enhances the accessibility and interoperability of LCA domain data and provides a scalable foundation for large-scale knowledge synthesis to support sustainability research.\u003c/p\u003e","manuscriptTitle":"From Literature to Knowledge Graphs: Automated Extraction and Retrieval of Life Cycle Assessment Data with Large Language Models","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-01-27 17:19:30","doi":"10.21203/rs.3.rs-8633241/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-04-09T13:36:04+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-09T22:00:27+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"242559425514762366328579509811865437240","date":"2026-02-09T08:06:47+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-02-03T20:59:33+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-01-21T16:03:24+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-01-21T12:05:25+00:00","index":"","fulltext":""},{"type":"submitted","content":"Journal of Industrial Ecology","date":"2026-01-18T19:02:12+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"journal-of-industrial-ecology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"44498","submissionUrl":"https://submission.springernature.com/new-submission/44498/3","title":"Journal of Industrial Ecology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"94386b72-ba97-49b6-a583-47baf658d344","owner":[],"postedDate":"January 27th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"in-revision","subjectAreas":[],"tags":[],"updatedAt":"2026-04-09T13:40:55+00:00","versionOfRecord":[],"versionCreatedAt":"2026-01-27 17:19:30","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8633241","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8633241","identity":"rs-8633241","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}