Multimodal LLM-Driven Forensic Framework for Criminal Intent Detection in Chat Histories

Multimodal LLM-Driven Forensic Framework for Criminal Intent Detection in Chat Histories | 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 Multimodal LLM-Driven Forensic Framework for Criminal Intent Detection in Chat Histories Aashi Manker, Juweriah Abdul Raheem Mohammed, Sairam Utukuru, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9360101/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract This paper proposes a multimodal forensic framework that processes audio, text, and image or video captions from suspected chats. Chat history play vital role in criminal case analysis to capture the hidden intent and slang of the people. Using the graph constructor model, we extract the entities involved in the discussion for building a social graph by identifying key influencers and the flow of discussion. A large language model analyzes the generated graphs to identify behavioral insights, motives, and individual influences on the discussion. A structured forensic report is generated to uncover the hidden intent, suspicious terms, and rank the influencers and key entities with confidence scores to support the investigators. Experimental analysis demonstrates that the generated final report is more accurate for the conclusion of the chat history. Artificial Intelligence and Machine Learning Digital Forensics Multimodal Analysis Large Language Models (LLM) Criminal Investigation Hidden Intent Detection Full Text Additional Declarations The authors declare no competing interests. Cite Share Download PDF Status: Posted Version 1 posted 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-9360101","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":622088033,"identity":"de4776d7-b287-4b61-bf18-ae237a978ed3","order_by":0,"name":"Aashi Manker","email":"","orcid":"","institution":"Chaitanya Bharathi Insitute of Technology","correspondingAuthor":false,"prefix":"","firstName":"Aashi","middleName":"","lastName":"Manker","suffix":""},{"id":622088155,"identity":"5f32fa0f-a94e-49eb-ba0d-490ba268ed75","order_by":1,"name":"Juweriah Abdul Raheem Mohammed","email":"","orcid":"","institution":"Chaitanya Bharathi Insitute of Technology","correspondingAuthor":false,"prefix":"","firstName":"Juweriah","middleName":"Abdul Raheem","lastName":"Mohammed","suffix":""},{"id":622088386,"identity":"feb5b63a-f6ec-44ce-94ec-6da434d47316","order_by":2,"name":"Sairam Utukuru","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA4ElEQVRIiWNgGAWjYFACxgYwZcDMwPgASPPwkaCFmdkApIWNaMsMGJjZJEAMglrk2w+3ffiZwyBvzs5/rPJrjp0MGwPzw0c38Bl+JrF5Zu82BsOdzcxst2W3JQMdxmZsnIPXPYnNDLzbGBIMDgO1SG5jBmrhYZPGp0W+/2Ez41+olmLJbfWEtTDcSGxmhtnC+HHbYcJaDG48bGaW3SYB8ouxNOO24zxszAT8It+f/pjx7TYbeXP+gw8//txWbc/P3vzwMV6HQQA4RhiYecAkYeUIwPiDFNWjYBSMglEwYgAAWmw+M7jXMywAAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0001-9639-2923","institution":"Chaitanya Bharathi Insitute of Technology","correspondingAuthor":true,"prefix":"","firstName":"Sairam","middleName":"","lastName":"Utukuru","suffix":""},{"id":622088515,"identity":"a4ab1852-9450-4fdd-b1fe-f8f0d72daebb","order_by":3,"name":"Siddhartha Mourya","email":"","orcid":"","institution":"Chaitanya Bharathi Insitute of Technology","correspondingAuthor":false,"prefix":"","firstName":"Siddhartha","middleName":"","lastName":"Mourya","suffix":""},{"id":622088719,"identity":"8c10bf3b-b1eb-482d-af43-27761374b7fa","order_by":4,"name":"Vivek Yelleti","email":"","orcid":"","institution":"SRM University AP","correspondingAuthor":false,"prefix":"","firstName":"Vivek","middleName":"","lastName":"Yelleti","suffix":""}],"badges":[],"createdAt":"2026-04-08 17:54:54","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-9360101/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9360101/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":106866867,"identity":"f3ad0d05-7075-4c45-933d-292c3fe76e8c","added_by":"auto","created_at":"2026-04-14 09:14:17","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2241237,"visible":true,"origin":"","legend":"","description":"","filename":"MPJournalPaperv4.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9360101/v1_covered_17281128-ce9b-4bca-9e02-f275b22325a8.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003eMultimodal LLM-Driven Forensic Framework for Criminal Intent Detection in Chat Histories\u003c/p\u003e","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Chaitanya Bharathi Institute of Technology","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Digital Forensics, Multimodal Analysis, Large Language Models (LLM), Criminal Investigation, Hidden Intent Detection","lastPublishedDoi":"10.21203/rs.3.rs-9360101/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9360101/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThis paper proposes a multimodal forensic framework that processes audio, text, and image or video captions from suspected chats. Chat history play vital role in criminal case analysis to capture the hidden intent and slang of the people. Using the graph constructor model, we extract the entities involved in the discussion for building a social graph by identifying key influencers and the flow of discussion. A large language model analyzes the generated graphs to identify behavioral insights, motives, and individual influences on the discussion. A structured forensic report is generated to uncover the hidden intent, suspicious terms, and rank the influencers and key entities with confidence scores to support the investigators. Experimental analysis demonstrates that the generated final report is more accurate for the conclusion of the chat history.\u003c/p\u003e","manuscriptTitle":"Multimodal LLM-Driven Forensic Framework for Criminal Intent Detection in Chat Histories","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-14 09:13:54","doi":"10.21203/rs.3.rs-9360101/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"399be581-66e9-4c63-96b7-b4c94f566a4f","owner":[],"postedDate":"April 14th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":66196363,"name":"Artificial Intelligence and Machine Learning"}],"tags":[],"updatedAt":"2026-04-14T09:13:54+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-14 09:13:54","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9360101","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9360101","identity":"rs-9360101","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}