Exploiting Distribution-Based Confidence Integration in Graph Neural Network Recommenders

Exploiting Distribution-Based Confidence Integration in Graph Neural Network Recommenders | 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 Exploiting Distribution-Based Confidence Integration in Graph Neural Network Recommenders Joel Machado Pires, Eduardo Ferreira da Silva, Frederico Araújo Durão This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8085564/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 Recommender systems assist users in navigating information-rich environments by delivering personalized content. While model-based collaborative filtering approaches, such as matrix factorization (MF) and graph neural networks (GNN), are widely adopted, the inherent uncertainty in user preferences and sparse data can lead to unreliable predictions. Confidence estimation has emerged as a strategy to quantify prediction reliability, yet its integration remains unexplored in GNN-based models, and prior methods often degrade accuracy or suffer from convergence issues. This study benchmarks four prominent confidence-aware models—OrdRec, Confidence-aware Probabilistic Matrix Factorization, Confidence-aware Bayesian Probabilistic Matrix Factorization, and Lightweight Beta Distribution across three public datasets: Amazon Movies and TVs, Jester Joke, and Movie Lens. We evaluate these models in terms of rating accuracy (root mean squared error), ranking quality (normalized discounted cumulative gain, mean average precision), and the quality of their confidence estimates. In addition, we propose a novel confidence-integrated model based on a deep graph attention network architecture. Experimental results reveal that while distribution-based confidence methods are highly sensitive to dataset characteristics and may harm accuracy, the proposed method demonstrates consistent performance across all datasets and metrics, outperforming prior distribution-based models. Nevertheless, challenges remain in aligning confidence estimates with prediction error. Artificial Intelligence and Machine Learning Information Retrieval and Management Applied Statistics recommendation reliability confidence in recommender systems distribution-based recommender models probabilistic recommender systems 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-8085564","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":543251094,"identity":"33fb8570-6e97-42d8-ac85-c31de6f09e53","order_by":0,"name":"Joel Machado Pires","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA1klEQVRIiWNgGAWjYBACxgYGxgMMBkAWewPDZyCVQIwWBogWngOMs4nSAgIHwKREApFamNubDxwuKLgnJz/zjWFzQQ1DnnkDIYf1HEs4PMOg2Njgdo5h84xjDMUyBwhpmZFjcJjHICFxg3SO+WMeNobEGYQcxjgj/wNIS/38mWcMm3n+EaUlhwGkJYHhBo9hM28bMVp6jhkA/ZJguOFMWmHzzD6JYglCWgzbmx8+LviTIC/ffnhjc8E3mzzCWhqAAY3EJ6iBgUGeAVXLKBgFo2AUjAJMAABz3EHxhu8gFgAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0002-8428-3516","institution":"Federal University of Bahia","correspondingAuthor":true,"prefix":"","firstName":"Joel","middleName":"Machado","lastName":"Pires","suffix":""},{"id":543251095,"identity":"e10fe2c2-fbf2-48f6-8394-acf756fc0b13","order_by":1,"name":"Eduardo Ferreira da Silva","email":"","orcid":"","institution":"Federal University of Bahia","correspondingAuthor":false,"prefix":"","firstName":"Eduardo","middleName":"Ferreira da","lastName":"Silva","suffix":""},{"id":543251096,"identity":"143bee07-9246-4256-97f6-b9a42206c625","order_by":2,"name":"Frederico Araújo Durão","email":"","orcid":"","institution":"Federal University of Bahia","correspondingAuthor":false,"prefix":"","firstName":"Frederico","middleName":"Araújo","lastName":"Durão","suffix":""}],"badges":[],"createdAt":"2025-11-11 10:21:01","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-8085564/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8085564/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":95800030,"identity":"35b5430e-1884-4c29-8112-e8f805336fad","added_by":"auto","created_at":"2025-11-13 08:21:24","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1342918,"visible":true,"origin":"","legend":"","description":"","filename":"JIIS2025Confidence.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8085564/v1_covered_8308241c-c9c8-4a57-af9c-761d82860958.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003eExploiting Distribution-Based Confidence Integration in Graph Neural Network Recommenders\u003c/p\u003e","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Federal University of Bahia","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":"recommendation reliability, confidence in recommender systems, distribution-based recommender models, probabilistic recommender systems","lastPublishedDoi":"10.21203/rs.3.rs-8085564/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8085564/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eRecommender systems assist users in navigating information-rich environments by delivering personalized content. While model-based collaborative filtering approaches, such as matrix factorization (MF) and graph neural networks (GNN), are widely adopted, the inherent uncertainty in user preferences and sparse data can lead to unreliable predictions. Confidence estimation has emerged as a strategy to quantify prediction reliability, yet its integration remains unexplored in GNN-based models, and prior methods often degrade accuracy or suffer from convergence issues. This study benchmarks four prominent confidence-aware models—OrdRec, Confidence-aware Probabilistic Matrix Factorization, Confidence-aware Bayesian Probabilistic Matrix Factorization, and Lightweight Beta Distribution across three public datasets: Amazon Movies and TVs, Jester Joke, and Movie Lens. We evaluate these models in terms of rating accuracy (root mean squared error), ranking quality (normalized discounted cumulative gain, mean average precision), and the quality of their confidence estimates. In addition, we propose a novel confidence-integrated model based on a deep graph attention network architecture. Experimental results reveal that while distribution-based confidence methods are highly sensitive to dataset characteristics and may harm accuracy, the proposed method demonstrates consistent performance across all datasets and metrics, outperforming prior distribution-based models. Nevertheless, challenges remain in aligning confidence estimates with prediction error.\u003c/p\u003e","manuscriptTitle":"Exploiting Distribution-Based Confidence Integration in Graph Neural Network Recommenders","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-12 03:35:17","doi":"10.21203/rs.3.rs-8085564/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":"1c9a899f-ea70-4ae1-bcc3-c14fb0171f57","owner":[],"postedDate":"November 12th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":57793241,"name":"Artificial Intelligence and Machine Learning"},{"id":57793242,"name":"Information Retrieval and Management"},{"id":57793243,"name":"Applied Statistics"}],"tags":[],"updatedAt":"2025-11-12T03:35:17+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-12 03:35:17","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8085564","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8085564","identity":"rs-8085564","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}