FlowSocial: A Dynamic Clustering-Based Federated Recommender System for Privacy-Preserving Social Media Personalization

FlowSocial: A Dynamic Clustering-Based Federated Recommender System for Privacy-Preserving Social Media Personalization | 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 FlowSocial: A Dynamic Clustering-Based Federated Recommender System for Privacy-Preserving Social Media Personalization Aryan Rajpurkar, Aaditya Malani, Advait Sankhe, Kriti Srivastava This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7248659/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 The spread of user-generated content on social media platforms has raised privacy concerns surrounding traditional centralized recommendation systems. This study presents FlowSocial, a unique dynamic clustering-based federated learning framework that addresses privacy preservation, scalability challenges, and personalization requirements. The system consists of client-side temporal-decay matrix factorization, giving only compact 32-dimensional user embeddings to maintain privacy while enabling collaborative learning. A key innovation is the post-aggregation dynamic learning mechanism which calculates the related user group after each round of federated learning which helps to capture important and evolving preference patterns without additional computer space. Moreover, FlowSocial exhibits rapid convergence: within the first five federated rounds, all key metrics improve sharply and then stabilize—demonstrating learning efficiency and scalability suitable for real-world, privacy-sensitive social media environments. These results establish FlowSocial as a compelling solution for privacy-preserving, multimodal recommendation systems deployed on social media platforms. Federated Learning Recommender Systems Dynamic Clustering Privacy Preservation Multimodal Embeddings Full Text Additional Declarations No competing interests reported. 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-7248659","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":499132262,"identity":"cb1f7346-9474-4d22-8900-94b9903d6e93","order_by":0,"name":"Aryan Rajpurkar","email":"","orcid":"","institution":"DJ Sanghvi College of Engineering","correspondingAuthor":false,"prefix":"","firstName":"Aryan","middleName":"","lastName":"Rajpurkar","suffix":""},{"id":499132263,"identity":"fdfd574c-1aad-473a-8d09-76a418ea2a3a","order_by":1,"name":"Aaditya Malani","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABE0lEQVRIiWNgGAWjYHACgwNgih2IEwz+yYHYBx4Q0gLWwwzSUnDAGKwlgYAWBrgWhg8HEhtAND4t/O2HNx7+2GYXzd/M/OzBA4M76fPDDj8E2mInp9uAXYvEmbSCAwfbknNnHGYzN0gweJa78XaaAVBLsrHZARzWHMgxAGphzm04zGAmkWDAnLtxdgJIy4HEbTi0yJ9/A9JSnzv/MPs3kJZ0w9npH/BqMbgBtuVw7obDPCBbDifIS+fgt8XwxrOCA2fOHc/deJinDKglzXCDdE7BgQQD3H6RO5+8+UNFWXXuvOPt2yR//LGRl5+dvvnDhwo7OZzeBwFGNmSnglUa4FEOBn+Q2PINhFSPglEwCkbBSAMA2a1u9NjKaAYAAAAASUVORK5CYII=","orcid":"","institution":"DJ Sanghvi College of Engineering","correspondingAuthor":true,"prefix":"","firstName":"Aaditya","middleName":"","lastName":"Malani","suffix":""},{"id":499132264,"identity":"cf310553-26db-450b-9225-caefb9ae7564","order_by":2,"name":"Advait Sankhe","email":"","orcid":"","institution":"DJ Sanghvi College of Engineering","correspondingAuthor":false,"prefix":"","firstName":"Advait","middleName":"","lastName":"Sankhe","suffix":""},{"id":499132265,"identity":"80d422e8-e5d8-4892-b50b-383f78c62ee3","order_by":3,"name":"Kriti Srivastava","email":"","orcid":"","institution":"DJ Sanghvi College of Engineering","correspondingAuthor":false,"prefix":"","firstName":"Kriti","middleName":"","lastName":"Srivastava","suffix":""}],"badges":[],"createdAt":"2025-07-30 05:23:57","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7248659/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7248659/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":96805482,"identity":"86fb2884-4d4c-4e2f-b00f-40686bf41cbd","added_by":"auto","created_at":"2025-11-26 09:10:06","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":589863,"visible":true,"origin":"","legend":"","description":"","filename":"FlowSocialspringerdiscoverartificialintelligence1.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7248659/v1_covered_300a60a3-ae96-4757-b0c7-e4a58f8e8103.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"FlowSocial: A Dynamic Clustering-Based Federated Recommender System for Privacy-Preserving Social Media Personalization","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"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":"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":"Federated Learning, Recommender Systems, Dynamic Clustering, Privacy Preservation, Multimodal Embeddings","lastPublishedDoi":"10.21203/rs.3.rs-7248659/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7248659/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"The spread of user-generated content on social media platforms has raised privacy concerns surrounding traditional centralized recommendation systems. This study presents FlowSocial, a unique dynamic clustering-based federated learning framework that addresses privacy preservation, scalability challenges, and personalization requirements. The system consists of client-side temporal-decay matrix factorization, giving only compact 32-dimensional user embeddings to maintain privacy while enabling collaborative learning. A key innovation is the post-aggregation dynamic learning mechanism which calculates the related user group after each round of federated learning which helps to capture important and evolving preference patterns without additional computer space. Moreover, FlowSocial exhibits rapid convergence: within the first five federated rounds, all key metrics improve sharply and then stabilize—demonstrating learning efficiency and scalability suitable for real-world, privacy-sensitive social media environments. These results establish FlowSocial as a compelling solution for privacy-preserving, multimodal recommendation systems deployed on social media platforms.","manuscriptTitle":"FlowSocial: A Dynamic Clustering-Based Federated Recommender System for Privacy-Preserving Social Media Personalization","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-14 07:30:22","doi":"10.21203/rs.3.rs-7248659/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":"849e1b80-6e58-4d24-8419-e90d21b33042","owner":[],"postedDate":"August 14th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-11-26T09:09:44+00:00","versionOfRecord":[],"versionCreatedAt":"2025-08-14 07:30:22","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7248659","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7248659","identity":"rs-7248659","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}