Dealing with Annotator Disagreement in Hate Speech Classification

Dealing with Annotator Disagreement in Hate Speech Classification | 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 Dealing with Annotator Disagreement in Hate Speech Classification Somaiyeh Dehghan, Mehmet Umut Sen, Berrin Yanikoglu This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7391084/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 Hate speech detection is a crucial task, especially on social media, where harmful content can spread quickly. Implementing machine learning models to automatically identify and address hate speech is essential for mitigating its impact and preventing its proliferation. The first step in developing an effective hate speech detection model is to acquire a high-quality dataset for training. Labeled data is essential for most natural language processing tasks, but categorizing hate speech is difficult due to the diverse and often subjective nature of hate speech, which can lead to varying interpretations and disagreements among annotators. This paper examines strategies for addressing annotator disagreement, an issue that has been largely overlooked. In particular, we evaluate various automatic approaches for aggregating multiple annotations, in the context of hate speech classification in Turkish tweets. Our work highlights the importance of the problem and provides state-of-the-art benchmark results for the detection and understanding of hate speech in online discourse. Data Annotation Annotator Disagreement Hate Speech Detection Natural Language Processing Large Language Models (LLMs) BERT Model 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-7391084","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":550110286,"identity":"3f5e3fc6-0a95-470c-b760-62b70b4a54bd","order_by":0,"name":"Somaiyeh Dehghan","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA+klEQVRIiWNgGAWjYJACCcYGhgQGBubjPz4AeWzsxGthS5CcAdLCTLwWHgNpHhCXkBb59gbGGx931Obxz25LMLb5tU2ej5mB8cPHHNxaDM4cYLaceeZ4scSdwweSc/tuG7YxMzBLztyGR4tEAps0b9uxxIYbaQmHc3tuMwK1sDHz4tEiPwOo5S9Qy/wbOYbNlj237QlqYbgB1MLYVpO44UaOMTPDj9uJBLUYnDnYbNnbdiBx4420NMbehtvJbcyMzXj9It/efPDGz7a6xHk3ko8x/Phz23Y+UOTDR3wOYwBGCgPDYSi7DS5CENRB6T/EKB4Fo2AUjIKRBgBC0lddt/U6sQAAAABJRU5ErkJggg==","orcid":"","institution":"Sabancı University","correspondingAuthor":true,"prefix":"","firstName":"Somaiyeh","middleName":"","lastName":"Dehghan","suffix":""},{"id":550110287,"identity":"a3a23761-a764-4306-a3dc-987312416e31","order_by":1,"name":"Mehmet Umut Sen","email":"","orcid":"","institution":"Sabancı University","correspondingAuthor":false,"prefix":"","firstName":"Mehmet","middleName":"Umut","lastName":"Sen","suffix":""},{"id":550110288,"identity":"f4b43609-36ba-44da-ab87-750cd587e341","order_by":2,"name":"Berrin Yanikoglu","email":"","orcid":"","institution":"Sabancı University","correspondingAuthor":false,"prefix":"","firstName":"Berrin","middleName":"","lastName":"Yanikoglu","suffix":""}],"badges":[],"createdAt":"2025-08-17 07:53:12","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7391084/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7391084/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":97019313,"identity":"1ea0c955-11e6-4c70-b111-3743c2e9494e","added_by":"auto","created_at":"2025-11-28 17:53:31","extension":"json","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5360,"visible":true,"origin":"","legend":"","description":"","filename":"31dcb9e79d23422d8b5950acbe773496.json","url":"https://assets-eu.researchsquare.com/files/rs-7391084/v1/767fc65ee37599dcc00be2b0.json"},{"id":102328976,"identity":"6c01e2e9-400d-473c-ab53-5c06832027b3","added_by":"auto","created_at":"2026-02-10 14:56:19","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":404702,"visible":true,"origin":"","legend":"","description":"","filename":"LRECannotationfocusp.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7391084/v1_covered_8a73597b-2ba9-4df4-872a-dfb7debf7f0c.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Dealing with Annotator Disagreement in Hate Speech Classification","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":"Data Annotation, Annotator Disagreement, Hate Speech Detection, Natural Language Processing, Large Language Models (LLMs), BERT Model","lastPublishedDoi":"10.21203/rs.3.rs-7391084/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7391084/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Hate speech detection is a crucial task, especially on social media, where harmful content can spread quickly. Implementing machine learning models to automatically identify and address hate speech is essential for mitigating its impact and preventing its proliferation. The first step in developing an effective hate speech detection model is to acquire a high-quality dataset for training. Labeled data is essential for most natural language processing tasks, but categorizing hate speech is difficult due to the diverse and often subjective nature of hate speech, which can lead to varying interpretations and disagreements among annotators. This paper examines strategies for addressing annotator disagreement, an issue that has been largely overlooked. In particular, we evaluate various automatic approaches for aggregating multiple annotations, in the context of hate speech classification in Turkish tweets. Our work highlights the importance of the problem and provides state-of-the-art benchmark results for the detection and understanding of hate speech in online discourse.","manuscriptTitle":"Dealing with Annotator Disagreement in Hate Speech Classification","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-28 17:53:20","doi":"10.21203/rs.3.rs-7391084/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":"6eab715d-7261-4d92-9c42-5e40f0fc83f8","owner":[],"postedDate":"November 28th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-02-10T14:55:15+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-28 17:53:20","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7391084","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7391084","identity":"rs-7391084","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}