Case study: the reliability of clustering solutions

Case study: the reliability of clustering solutions | 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 Case study: the reliability of clustering solutions Mustafa Kadhim, Guangxi Lu, Jianbo Wang, Zhao Kang, Ruyi Tang, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4690535/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 Unsupervised clustering analysis (CA) models are crucial in various machine learning tasks aiming to discover the data structure and hidden patterns between samples. The performance of CA is affected by overlapping issue (samples that are similar to each other but do not belong to the same cluster). These samples may occur when the data grows over time, which causes the need to redefine features , the distance matrix, centroids, or other parameters. As recent CA models achieved optimal solutions, the details of parameters have been ignored due to the “no-free lunch” theory, which raises the question about their reliability. This study conducts an in-depth investigation of the complex relationships among the given issues, aiming to clarify the key problems associated with unsupervised models that may lead to inaccurate prediction of patterns, inappropriate partitioning , and false confidence in outcomes. We designed a novel experiment* that produced over 271 million possible settings based on various models, equations, and procedures. The experimental results* demonstrate that obtaining optimal parameters for optimal solutions is as challenging as finding a needle in a haystack. Finally, we designed an open issue* that challenges the CA models because our experiments raised doubts over the use of supervised information upfront to determine the optimal solutions, which is against the concept of CA. Pattern Recognition Exclusive Clustering Data Overlapping Machine Learning Feature Selections Full Text Additional Declarations No competing interests reported. Supplementary Files Supplementarymaterial.zip 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-4690535","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":346594632,"identity":"cb1dd653-b51b-409d-b02e-8a133c4d6c13","order_by":0,"name":"Mustafa Kadhim","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABCElEQVRIie3NMWvCQBTA8VcCcQm4Xgj2MwQCQbD4RbrkcaBL2rE6OGRKN7PGb+GUtXccOKW9NeBSKXTqkKnooh6tQSindnS4Pxy8B+/HAZhMVxgBCCFqtjHcNaP1H6KOShg01xcIHIm4TNxnUbyvYrj3peQ1m0jM2mqAcQ+T1ivTEc8ZjHws4GFeUYuwxRJnuRqgHGLiPEY6cgtxSH6JBV5tL1ENNtykAhPi+FrS/joQKaw1277hixSKbE8TjzS/MGoTnjKcA1UkOU3c/PPph8wqGnb5lAZ5RQMSLYZB6sRaQiQt3I0iU8k/Kvbd72QZX9X1pNfJWqWWHNr92SP17DP3JpPJZDrfHh06ZI0s5GYsAAAAAElFTkSuQmCC","orcid":"","institution":"University of Electronic Science and Technology of China","correspondingAuthor":true,"prefix":"","firstName":"Mustafa","middleName":"","lastName":"Kadhim","suffix":""},{"id":346594633,"identity":"94973ce8-0aec-4e0a-bfde-09dc0176f35b","order_by":1,"name":"Guangxi Lu","email":"","orcid":"","institution":"University of Electronic Science and Technology of China","correspondingAuthor":false,"prefix":"","firstName":"Guangxi","middleName":"","lastName":"Lu","suffix":""},{"id":346594634,"identity":"1dcfd44f-7629-4041-a355-9a1307173787","order_by":2,"name":"Jianbo Wang","email":"","orcid":"","institution":"University of Electronic Science and Technology of China","correspondingAuthor":false,"prefix":"","firstName":"Jianbo","middleName":"","lastName":"Wang","suffix":""},{"id":346594635,"identity":"1afc23b4-99b2-40c1-a327-241c97f1dad1","order_by":3,"name":"Zhao Kang","email":"","orcid":"","institution":"University of Electronic Science and Technology of China","correspondingAuthor":false,"prefix":"","firstName":"Zhao","middleName":"","lastName":"Kang","suffix":""},{"id":346594636,"identity":"6ed227c5-eae8-42fc-950f-f0a047ffece0","order_by":4,"name":"Ruyi Tang","email":"","orcid":"","institution":"Phoenix Intelligent Innovation Technology Co., Ltd, Shapingba Industrial Zone A","correspondingAuthor":false,"prefix":"","firstName":"Ruyi","middleName":"","lastName":"Tang","suffix":""},{"id":346594637,"identity":"05f6b07b-10f5-430f-a22e-3ede960f62ce","order_by":5,"name":"Songlin Ye","email":"","orcid":"","institution":"Phoenix Intelligent Innovation Technology Co., Ltd, Shapingba Industrial Zone A","correspondingAuthor":false,"prefix":"","firstName":"Songlin","middleName":"","lastName":"Ye","suffix":""}],"badges":[],"createdAt":"2024-07-05 07:51:16","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4690535/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4690535/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":63600751,"identity":"a075d1e9-b57e-42a8-9a73-287dbd3017d6","added_by":"auto","created_at":"2024-08-30 05:21:00","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":8568313,"visible":true,"origin":"","legend":"","description":"","filename":"Manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4690535/v1_covered_6a3bb50b-4f58-4ce1-97e4-319a7f8552be.pdf"},{"id":63599828,"identity":"45c49f48-c781-4006-98e8-0dc21061f5fd","added_by":"auto","created_at":"2024-08-30 05:04:51","extension":"zip","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":166150829,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementarymaterial.zip","url":"https://assets-eu.researchsquare.com/files/rs-4690535/v1/d88d31a8213a7e10c12f99df.zip"}],"financialInterests":"No competing interests reported.","formattedTitle":"Case study: the reliability of clustering solutions","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"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":"Pattern Recognition, Exclusive Clustering, Data Overlapping, Machine Learning, Feature Selections","lastPublishedDoi":"10.21203/rs.3.rs-4690535/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4690535/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eUnsupervised clustering analysis (CA) models are crucial in various machine learning tasks aiming to discover the data structure and hidden patterns between samples. The performance of CA is affected by overlapping issue (samples that are similar to each other but do not belong to the same cluster). These samples may occur when the data grows over time, which causes the need to redefine features , the distance matrix, centroids, or other parameters. As recent CA models achieved optimal solutions, the details of parameters have been ignored due to the “no-free lunch” theory, which raises the question about their reliability. This study conducts an in-depth investigation of the complex relationships among the given issues, aiming to clarify the key problems associated with unsupervised models that may lead to inaccurate prediction of patterns, inappropriate partitioning , and false confidence in outcomes. We designed a novel experiment* that produced over 271 million possible settings based on various models, equations, and procedures. The experimental results* demonstrate that obtaining optimal parameters for optimal solutions is as challenging as finding a needle in a haystack. Finally, we designed an open issue* that challenges the CA models because our experiments raised doubts over the use of supervised information upfront to determine the optimal solutions, which is against the concept of CA.\u003c/p\u003e","manuscriptTitle":"Case study: the reliability of clustering solutions","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-08-30 05:04:41","doi":"10.21203/rs.3.rs-4690535/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":"63f64197-5405-46f3-be40-653244f9108e","owner":[],"postedDate":"August 30th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2024-08-30T05:04:41+00:00","versionOfRecord":[],"versionCreatedAt":"2024-08-30 05:04:41","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4690535","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4690535","identity":"rs-4690535","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}