RLIM: Representation Learning Method for Influence Maximization in social networks

RLIM: Representation Learning Method for Influence Maximization in social networks | 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 RLIM: Representation Learning Method for Influence Maximization in social networks Sun Chengai, Duan Xiuliang, Qiu Liqing, Shi Qiang, Li Tengteng This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-381918/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 A core issue in influence propagation is influence maximization, which aims to find a group of nodes under a specific information diffusion model and maximize the final influence of this group of nodes. The limitation of the existing researches is that they excessively depend on the information diffusion model and randomly set the propagation ability (probability). Therefore, most of the algorithms for solving the influence maximization problem are basically difficult to expand in large social networks. Another challenge is that fewer researchers have paid attention to the problem of the large difference between the estimated influence spread and the actual influence spread. A measure to solve the influence maximization problem is applying advanced neural network architecture also represents learning method. Based on this idea, the paper proposes Representation Learning for Influence Maximization (RLIM) algorithm. The premise of this algorithm is to construct the influence cascade of each source node. The key is to adopt neural network architecture to realize the prediction of propagation ability. The purpose is to apply the propagation ability to the influence maximization problem by representation learning. Furthermore, the results of the experiments show that RLIM algorithm has greater diffusion ability than the state-of-the-art algorithms on different online social network data sets, and the diffusion of information is more accurate. Computational Biology Bioinformatics Computer Architecture and Engineering Influence maximization information diffusion model propagation probability neural network architecture representation learning Figures Figure 1 Figure 2 Figure 3 Figure 4 Full Text Due to technical limitations, full-text HTML conversion of this manuscript could not be completed. However, the manuscript can be downloaded and accessed as a PDF. 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-381918","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":20239448,"identity":"0d1412e3-78a3-4ecd-9708-078aa2fb9984","order_by":0,"name":"Sun Chengai","email":"","orcid":"","institution":"Shandong University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Sun","middleName":"","lastName":"Chengai","suffix":""},{"id":20239451,"identity":"c7ff041f-2441-43db-bec5-e228f4ad2097","order_by":1,"name":"Duan Xiuliang","email":"","orcid":"","institution":"Shandong University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Duan","middleName":"","lastName":"Xiuliang","suffix":""},{"id":20239452,"identity":"8ed51331-6c52-4786-abeb-5e3279ed1baa","order_by":2,"name":"Qiu Liqing","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA70lEQVRIiWNgGAWjYDACZgbGAwkMEgwM7I2NDxIqJOTkidDCANHCc/iwwYczFsaGDURYdABMSqSlSc5sq0iEcnEDvuM8BgcelFnkyTvkGEjzzpNIYGxgfvjoBh4tkoeBWhLOSRQbHjhjYMy7TSKPnYHN2DgHjxYDkJbENonEjY09BslALcWMDTxs0sRpaeYxOMw7RyKx4QCxWuazsSU2zmwgQovkYbYCkF8SN/AwH2b4cEzC2LCZgF/4zh/e+PBHWV3i/PkP238k1NTJybM3P3yMTwskFtiALjwAE2HGpxxZi3wDIZWjYBSMglEwYgEAj1dRu94DCskAAAAASUVORK5CYII=","orcid":"","institution":"Shandong University of Science and Technology","correspondingAuthor":true,"prefix":"","firstName":"Qiu","middleName":"","lastName":"Liqing","suffix":""},{"id":20239454,"identity":"add22c3c-5fe0-4a53-adee-e8f6e7bbfa40","order_by":3,"name":"Shi Qiang","email":"","orcid":"","institution":"Shandong University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Shi","middleName":"","lastName":"Qiang","suffix":""},{"id":20239455,"identity":"1aae1931-1d40-4e57-960b-2daeff149072","order_by":4,"name":"Li Tengteng","email":"","orcid":"","institution":"Shandong University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Li","middleName":"","lastName":"Tengteng","suffix":""}],"badges":[],"createdAt":"2021-04-01 03:44:01","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-381918/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-381918/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":7799852,"identity":"5b53b523-8347-45b3-9c47-7d225379341e","added_by":"auto","created_at":"2021-04-08 16:02:19","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":83752,"visible":true,"origin":"","legend":"(a) The forwarding of information is carried out within a time step represented by a natural number. (b) The positions of nodes in the network are relative, and nodes in the same circle have the same level of influence probability. (c) The time interval for the information to be forwarded to neighboring nodes is a continuous random variable. (d) The information flow between nodes is not fixed, and the propagation direction is determined by P and Q.","description":"","filename":"fig1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-381918/v1/d5e2c705373f705116d6f71b.jpg"},{"id":7799853,"identity":"ff816c1d-9314-4f95-94e2-0dd4abf0528c","added_by":"auto","created_at":"2021-04-08 16:02:19","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":57461,"visible":true,"origin":"","legend":"Representation learning method framework.","description":"","filename":"fig2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-381918/v1/f9dacbb4fcef92b21e35efa8.jpg"},{"id":7799851,"identity":"767111c8-c6f4-4ef6-9666-8ee5f70e17f7","added_by":"auto","created_at":"2021-04-08 16:02:19","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":50280,"visible":true,"origin":"","legend":"Visualization results of the part dataset of Digg network. Each participating user is mapped to the 2-D space using t-SNE package. The colors of the nodes refer to different types of users.","description":"","filename":"fig3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-381918/v1/84bae801cae72951b1696502.jpg"},{"id":7799373,"identity":"165c38c3-da2c-497e-b61b-3623a00b096e","added_by":"auto","created_at":"2021-04-08 15:59:19","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":77626,"visible":true,"origin":"","legend":"The influence spread of different algorithms on the four OSNs.","description":"","filename":"fig4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-381918/v1/1329ca270476b388aba4eeff.jpg"},{"id":13618491,"identity":"f635c639-209a-426b-9cda-5c725105bd48","added_by":"auto","created_at":"2021-09-17 06:57:14","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":720071,"visible":true,"origin":"","legend":"","description":"","filename":"RLIM.pdf","url":"https://assets-eu.researchsquare.com/files/rs-381918/v1_covered.pdf"},{"id":9508083,"identity":"ca3e53de-1c65-4c9b-a0f0-497bf80c8d07","added_by":"auto","created_at":"2021-05-24 13:59:19","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":716457,"visible":true,"origin":"","legend":"","description":"","filename":"RLIM.pdf","url":"https://assets-eu.researchsquare.com/files/rs-381918/v1_covered.pdf"},{"id":7800242,"identity":"b44cb38a-13bc-4cdb-b0f9-61950aacd2ab","added_by":"auto","created_at":"2021-04-08 16:05:24","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":597417,"visible":true,"origin":"","legend":"","description":"","filename":"RLIM.pdf","url":"https://assets-eu.researchsquare.com/files/rs-381918/v1_stamped.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"RLIM: Representation Learning Method for Influence Maximization in social networks","fulltext":[{"header":"Full Text","content":"\u003cp\u003eDue to technical limitations, full-text HTML conversion of this manuscript could not be completed. However, the manuscript can be downloaded and accessed as a PDF.\u003c/p\u003e"}],"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":false,"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":"Influence maximization, information diffusion model, propagation probability, neural network architecture, representation learning","lastPublishedDoi":"10.21203/rs.3.rs-381918/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-381918/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"A core issue in influence propagation is influence maximization, which aims to find a group of nodes under a specific information diffusion model and maximize the final influence of this group of nodes. The limitation of the existing researches is that they excessively depend on the information diffusion model and randomly set the propagation ability (probability). Therefore, most of the algorithms for solving the influence maximization problem are basically difficult to expand in large social networks. Another challenge is that fewer researchers have paid attention to the problem of the large difference between the estimated influence spread and the actual influence spread. A measure to solve the influence maximization problem is applying advanced neural network architecture also represents learning method. Based on this idea, the paper proposes Representation Learning for Influence Maximization (RLIM) algorithm. The premise of this algorithm is to construct the influence cascade of each source node. The key is to adopt neural network architecture to realize the prediction of propagation ability. The purpose is to apply the propagation ability to the influence maximization problem by representation learning. Furthermore, the results of the experiments show that RLIM algorithm has greater diffusion ability than the state-of-the-art algorithms on different online social network data sets, and the diffusion of information is more accurate.","manuscriptTitle":"RLIM: Representation Learning Method for Influence Maximization in social networks","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2021-04-08 15:59:17","doi":"10.21203/rs.3.rs-381918/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":"b02e8b5e-c9aa-41c6-8ac1-2ece67de7069","owner":[],"postedDate":"April 8th, 2021","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":3497337,"name":"Computational Biology"},{"id":3497338,"name":"Bioinformatics"},{"id":3497339,"name":"Computer Architecture and Engineering"}],"tags":[],"updatedAt":"2021-05-24T13:59:07+00:00","versionOfRecord":[],"versionCreatedAt":"2021-04-08 15:59:17","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-381918","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-381918","identity":"rs-381918","version":["v1"]},"buildId":"_2-kVJe1T_tPrBINL-cwx","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}