MetaCorpus

A meta corpus of social media corpus. Part of the SocialMediaIE Project.

• MetaCorpus
  • Twitter
    • Classification
    • Stance detection
    • Tagging
      • NER datasets
    • Entity Linking
    • Machine Translation
    • Paraphrase identification
    • Rumour detection
    • Treebank and parsing
    • Question answering
    • Conversations
    • Information Retrieval
    • Multimodal
    • Sentence Similarity
    • Summarization
    • General
    • Tools, Tips, and Tricks
    • Embeddings
    • Unlabled topic specific data dumps
  • Facebook
    • Classification
  • Youtube:
    • Classification
    • Videos
    • General
  • Reddit
    • Classification
    • Summarization
    • Sequence Tagging
    • Named Entitis
    • Conversations
    • Tools
  • Gab
    • Summarization
    • Amazon
  • About.me
  • Whatsapp
  • Delicious
  • Ask.fm
  • Flickr
    • Popularity prediction
  • Conversations
  • News Comments
  • Weibo
  • Whisper
  • SMS
  • General

Table of contents generated with markdown-toc

Twitter

Classification

• Science-related speech detection (CIKM2022) - Dataset/Model, Paper
• Hate and abusive speech - https://github.com/ENCASEH2020/hatespeech-twitter
• HateEval (SemEval 2019) - https://github.com/cicl2018/HateEvalTeam
• Aggregation identification - https://github.com/SilentFlame/AggressionDetection
• Hate speech ICWSM 2017 - https://github.com/t-davidson/hate-speech-and-offensive-language
• Moral Foundations Twitter Corpus - https://psyarxiv.com/w4f72 [JSON data]
• Vulgarity prediction on tweets - https://github.com/ericholgate/vulgartwitter
• CrisisNLP - http://crisisnlp.qcri.org
• CrisisLex - http://crisislex.org
• CREDBANK - credible events for tweets - http://compsocial.github.io/CREDBANK-data/
• Social roles in Twitter - https://data.csiro.au/dap/landingpage?pid=csiro:33845
• Biomedical entities in tweets - http://diego.asu.edu/Publications/ADRMine.html
• UK election sentiment - https://figshare.com/articles/EACL_2017_-_Multi-target_UK_election_Twitter_sentiment_corpus/4479563
• Novelty Detection in Tweets - https://github.com/CrowdTruth/Novelty_Detection Project Page
• MovieTweetings - https://github.com/sidooms/MovieTweetings
• Monthly twitter Archive by the Internet Archive team from 2012-2018 - https://archive.org/details/twitterstream?&sort=-date&page=1
• EmoContext dataset archived at Wayback Machine - Train, Dev, Test
• Multi target sentiment - https://github.com/bluemonk482/tdparse
• RumourEval 2019 Data
• Sentiment, Emotion, Purpose, and Style in Electoral Tweets and Semantic Role Labeling of Emotions in Tweets
• Hashtag Emotion Corpus (aka Twitter Emotion Corpus, or TEC)
• List of sentiment corpora: https://www.w3.org/community/sentiment/wiki/Datasets
• Code Mixing and POS between English, Hindi, Bengali, and Telgu for FB, Twitter, and Whatsapp: http://www.amitavadas.com/Code-Mixing.html
• Twitter topic communities: https://github.com/uwnlp/twittercommunities
• Many useful datasets by Arkaitz Zubiaga: http://www.zubiaga.org/datasets/
• Twitter Optimism Dataset - http://lit.eecs.umich.edu/downloads.html#Twitter%20Optimism%20Dataset
• Grounded Emotions - http://web.eecs.umich.edu/~mihalcea/downloads/GroundedEmotions.tar.gz
• Vulgar tweet dataset - https://github.com/ericholgate/VulgarFunctionsTwitter/blob/master/Vulgar_Functions_Dataset.tsv
• Misogynistic tweet classification in Italian and English - https://amievalita2018.wordpress.com/
• Hatespeech detection on Facebook and Twitter (Italian) - http://www.di.unito.it/~tutreeb/haspeede-evalita18/index.html
• How to build a twitter archive - https://arxiv.org/abs/1611.08144
• Internet Archive twitter stream - https://archive.org/details/twitterstream
• Twitter datasets collected for specific events, e.g. US elections, has doi for each - https://tweetsets.library.gwu.edu/
• Tweets tagged with geolocations in geonames - https://github.com/geovista/GeoCorpora/blob/master/geocorpora_1544784178012.tsv
• Tweet geolocation datasets - https://github.com/afshinrahimi/geographconv
• Tweets tagged pro versus anti vaccine - http://www.cs.jhu.edu/~mdredze/data/
• Twitter Firstname, Lastname clusters - http://www.clsp.jhu.edu/~sbergsma/TwitterClusters/ (can be used for name location prediction)
• Twitter user level datasets for political ideologies - http://www.preotiuc.ro/resources.html
• A collection of tweet datasets - https://data.world/datasets/twitter
• End of Term 2016 U.S. Government Twitter Archive - https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/TQBLWZ
• Factuality of tweets - https://github.com/jacobeisenstein/twitter-certainty
• Parallel data for Machine Translation from tweets - http://www.cs.cmu.edu/~lingwang/microtopia/#translation
• Tweet to news linked corpora - http://www.cs.columbia.edu/~weiwei/code/acl2013.zip Source: http://www.cs.columbia.edu/~weiwei/code.html
• Congressional tweets labeled for frames and issues - https://github.com/kmjohnson/twitter-framing
  • train test splits can be used from https://github.com/coastalcph/issue_framing/tree/master/data/congressional_tweets
  • Same data labeled for moral foundations labels - https://github.com/kmjohnson/twitter-morals
• Presidential candidate tweets labeled for issues - https://github.com/kmjohnson/political-twitter
• Daily update of US congress politician tweets - https://github.com/alexlitel/congresstweets/tree/master/data
• Salience in News Tweets - http://data.crowdtruth.org/salience-news-tweets/ https://zenodo.org/record/46477
• Routinely collected Tweets datasets on topics of importance in US (esp US political tweets) - https://www.docnow.io/catalog/
• Twitter Cascades data - https://zenodo.org/record/2585375#.XVrLIOhKh3g
• Twitter News dataset - https://users.dcc.uchile.cl/~mquezada/breakingnews/
• 30 events dataset - https://figshare.com/articles/Twitter_event_datasets_2012-2016_/5100460
• Open Knowledge Representation for twitter events - http://u.cs.biu.ac.il/~nlp/resources/downloads/twitter-events/
• All verified Twitter users - https://www.reddit.com/r/datasets/comments/8s6nqz/all_verified_twitter_users_100_complete_in_ndjson/
• Many twitter datasets by pushshift - https://files.pushshift.io/
• Real time trending topic classification - http://nlp.uned.es/~damiano/datasets/TT-classification.html
• Balancing information exposure in social networks - https://github.com/gvrkiran/BalancedExposure - https://users.ics.aalto.fi/kiran/BalanceExposure/
• Online public shaming - https://zenodo.org/record/2587843#.XVrXzOhKh3g
• Abusive behavior - https://zenodo.org/record/1216127#.XVriSehKh3g
• GermEval offence evaluation task - https://projects.fzai.h-da.de/iggsa/projekt/
• GermEval-2018-Data - https://github.com/uds-lsv/GermEval-2018-Data * OffenseEval - https://competitions.codalab.org/competitions/20011
• Frame Semantics Twitter (AAAI 2015) - https://bitbucket.org/lowlands/release/src/master/AAAI15/ (Paper: https://www.aaai.org/ocs/index.php/AAAI/AAAI15/paper/view/9349)
• SemEval 2017 dataset Gold - Task Download
• MTSA - Multiannotated tweet sentiment dataset - https://www.aclweb.org/anthology/N18-1171/ from Sentiment Analysis: It’s Complicated! - https://github.com/networkdynamics/mcgill-tsa
• Freebase Annotations for TRAC KBA challenge (includes twitter stream data) - http://aws-publicdatasets.s3.amazonaws.com/trec/kba/FAKBA1/index.html
• Twitter Machine Translation corpus - http://komunitatea.elhuyar.eus/tweetmt/resources/
• Twitter Language Identification - http://komunitatea.elhuyar.eus/tweetlid/
• Twitter classification using open directory project categories - http://www.zubiaga.org/datasets/odptweets/
• Twitter Trending Topic - http://nlp.uned.es/~damiano/datasets/TT-classification.html
• Various Twitter datasets for personality classification as well as MSM2013 entity linking data - https://web.sas.upenn.edu/danielpr/resources/
• Weibo and Twitter data with 5,000 claims that scale to five million relevant microblog posts available at: http://alt.qcri.org/~wgao/data/rumdect.zip Paper
• Open domain targeted sentiment - https://web.archive.org/web/20200427155239/http://www.m-mitchell.com/code/
• Tweet span sentiment identification - http://alt.qcri.org/semeval2014/task9/index.php?id=data-and-tools
• Twitter Stereotype Extraction - https://github.com/kennyjoseph/twitter_stereotype_extraction
• Twitter Identity Extraction
• Political Parody (politican tweets and their parody tweets) - https://archive.org/details/parody_data_acl20
• Italian Twitter Corpus of Hate Speech - https://github.com/msang/hate-speech-corpus
• TRAC 2020 Multilingual Annotated Corpus of Misogyny and Aggression - https://sites.google.com/view/trac2/shared-task?authuser=0 - drive
• Detecting Nastiness in Social Media - http://ritual.uh.edu/wp-content/uploads/2017/08/ask.fm_dataset.zip
• Social Bias Frames (Twitter, Reddit, etc for multiple tags) - https://homes.cs.washington.edu/~msap/social-bias-frames/
• User level abusive, hate, and bully classification - https://zenodo.org/record/1184178#.Xgu24UdKh3g
• EveTAR test collection, Arabic Test Collection for multiple information retrieval tasks in Twitter (Event detection, Ad-hoc search, Timeline generation, Real-time summarization) - http://qufaculty.qu.edu.qa/telsayed/evetar/ . It supports:
• MeToo Dataset annotated for Text_Only_Informative,Image_Only_Informative,Directed_Hate,Generalized_Hate,Sarcasm,Allegation,Justification,Refutation,Support,Oppose - https://github.com/akash418/MeTooMMD
• Arabic HateSpeech annotated corpora - https://github.com/nuhaalbadi/Arabic_hatespeech
• Arabic Twitter Bots annotated user data - https://github.com/nuhaalbadi/ArabicBots
• Social Media Mining for Health Applications (Classification, Tagging, and Linking) - https://healthlanguageprocessing.org/smm4h/challenge/ CodaLab
• Sentiment and Entity Corpus on Climate Change tweets - https://gate.ac.uk/projects/decarbonet/datasets.html
• Twitter Dialogue Corpus - https://github.com/Phylliida/Dialogue-Datasets
• HAHA - Humor Analysis based on Human Annotation, a task to classify tweets in Spanish as humorous or not, and to determine how funny they are - https://www.fing.edu.uy/inco/grupos/pln/haha/index.html#data
• Multiple Hate speech datasets - https://github.com/leondz/hatespeechdata
• SSIX BREXIT Twitter Annotated Data Set (sentiment) - https://zenodo.org/record/1229649#.XyNZU_hKidY https://bitbucket.org/ssix-project/brexit-gold-standard/src/master/
• HaterNet a system for detecting and analyzing hate speech in Twitter (Spanish) - https://zenodo.org/record/2592149#.XyNbTPhKidY
• Webis Clickbait Corpus 2016 (Webis-Clickbait-16) (2992 tweets) - https://zenodo.org/record/3251557#.XyNcgvhKidY
• Webis Clickbait Corpus 2017 (Webis-Clickbait-17) (38,517 tweets) - https://zenodo.org/record/3346491#.XyNd3_hKidY - Clickbait Challenge
• RepLab Summarization Dataset (also topic labeling) - https://zenodo.org/record/2536801#.XyNdcvhKidY
• FIRE 2018 IRMiDis track dataset: Fact-checkable tweets posted during disasters - https://zenodo.org/record/3336477#.XyNiIPhKidY
• Political leaning of Twitter Users based on list membership (includes 3 months tweets) - https://github.com/klout/opendata/blob/master/political_leaning/README.md
• The UMass Global English on Twitter Dataset - https://www.kaggle.com/rtatman/the-umass-global-english-on-twitter-dataset (This dataset contains tweets from a variety of languages, tagged for whether they are in English or not, whether they contain code-switching, whether they includes names from a different language and whether they were generated automatically.)
• Twitter Sentence Specificity - https://github.com/wjko2/Domain-Agnostic-Sentence-Specificity-Prediction/tree/master/dataset/data
• Sarcasm Detection (Shared Task 2nd FigLang Workshop at ACL 2020) - https://github.com/EducationalTestingService/sarcasm
• Webis Clickbait Corpus 2017 - https://webis.de/data/webis-clickbait-17
• Tweet Wikipedia Entity Prediction - https://github.com/lunafeng/ELTDS
• Twitter data annotated with topic labels - http://www.site.uottawa.ca/~diana/resources/
• Twitter data with annotated location expressions at city, state/province, and country level - http://www.site.uottawa.ca/~diana/resources/
• Language specificity in Tweets - https://github.com/cs329yangzhong/specificityTwitter
• Geolocation prediction from text - https://archive.org/details/twitter_cikm_2010
• Social Bias Inference Corpus (SBIC) contains 150k structured annotations of social media posts - https://homes.cs.washington.edu/~msap/social-bias-frames/
• Civil Unrest Tweets - https://github.com/AADeLucia/JHU-CUT
• Japanese Tweet Sentiment Corpus (500k) - http://www.db.info.gifu-u.ac.jp/data/Data_5d832973308d57446583ed9f
• Japanese Covid 19 Tweet Relevance - http://www.db.info.gifu-u.ac.jp/data/Data_5f02db873363f976fce930d1
• Arabic Tweet Sentiment - https://archive.ics.uci.edu/ml/datasets/Twitter+Data+set+for+Arabic+Sentiment+Analysis
• English Spanish Tweet Hate Speech - https://competitions.codalab.org/competitions/19935
• TREC-IS - Detailed annotated Crisis Tweets - http://dcs.gla.ac.uk/~richardm/TREC_IS/2020/data.html
• Hinglish Spanglish Code-Mixed Tweet Sentiment - https://ritual-uh.github.io/sentimix2020/res
• Manually annotated Arabic Sentiment data - https://github.com/motazsaad/arabic-sentiment-analysis
• Semeval 2017 English Arabic Sentiment data - https://alt.qcri.org/semeval2017/task4/index.php?id=results
• Italian tweets factual annotation - https://github.com/evalita2016/data
• Tweet Image to Emoji dataset (4M) - https://www.cs.utexas.edu/~ziad/emoji_visual_sentiment.html
• Visual Public Emotions dataset - https://github.com/cvlab-stonybrook/Emotion-Prediction/tree/master/main/PublicEmotionDatasets
• EmotionGIF 2020 Shared task on tweet text to GIF category prediction - https://sites.google.com/view/emotiongif-2020/shared-task/dataset?authuser=0
• iSarcasm - sarcasm, irony, satire, understatement, overstatement, rhetorical question - https://github.com/silviu-oprea/iSarcasm
• Irony Sarcasm Analysis Corpus (includes geolocation info) - http://romanklinger.de/ironysarcasm/
• Tweet text classification benchmark - https://github.com/cardiffnlp/tweeteval
• Unified Multilingual Sentiment Analysis Benchmark - https://github.com/cardiffnlp/xlm-t
• Feel-IT - Italian Tweet Sentiment and Emotion - https://towardsdatascience.com/sentiment-analysis-and-emotion-recognition-in-italian-using-bert-92f5c8fe8a2
• Multilingual Twitter Hatespeech with Demographic Attributes - https://github.com/xiaoleihuang/Multilingual_Fairness_LREC
• Gender and Individual/Org predictor from Tweet User - https://bitbucket.org/mdredze/demographer
• Hurricane Emotions - https://github.com/shreydesai/hurricane
• COVID19 Fake News Detection in English - https://competitions.codalab.org/competitions/26655 - https://constraint-shared-task-2021.github.io/
• Hostile Post Detection in Hindi - https://competitions.codalab.org/competitions/26654 - https://constraint-shared-task-2021.github.io/
• Implicit and Explicit Hate - https://github.com/GT-SALT/implicit-hate
• SMILE: Twitter Emotion Classification using Domain Adaptation - https://www.kaggle.com/ashkhagan/smile-twitter-emotion-dataset
• Sarcasm Detection based on User and Reader - https://github.com/bshmueli/SPIRS
• Arabic Tweet Sentiment - https://github.com/komari6/Arabic-twitter-corpus-AJGT Parsed format on Huggingface Dataset
• Tweets Hate Speech Detection - https://huggingface.co/datasets/tweets_hate_speech_detection
• TRACT: Tweets Reporting Abuse Classification Task Corpus - https://data.mendeley.com/datasets/my2vkfyffd/2
• TM-Senti: large-scale, multilingual and longitudinal Twitter sentiment dataset sampled through distant supervision from the Twitter Stream Grab archive (2013 - 2020, 7 langs) - https://figshare.com/articles/dataset/TM-Senti/16438281
• DynaSent: Dynamic Sentiment Analysis Dataset - https://github.com/cgpotts/dynasent
• ICON2021 Shared Task on Multilingual Gender Biased and Communal Language Identification - https://competitions.codalab.org/competitions/35482
• COMRADES Crowdsourced Informativeness Dataset (CCSID) - https://figshare.com/articles/dataset/COMRADES_Crowdsourced_Informativeness_Datased_CCSID_/5787693
• Hate Speech Dataset - https://github.com/Mainack/hatespeech-data-HT-2017
• Estimating Ideological Positions with Twitter Data - https://github.com/pablobarbera/twitter_ideology
• SemEval-2018 task 3 - Irony detection in English tweets - https://competitions.codalab.org/competitions/17468 - Data Download - https://github.com/Cyvhee/SemEval2018-Task3
• SemEval-2018 Task 1: Affect in Tweets (AIT-2018) - https://competitions.codalab.org/competitions/17751 - Data Download - http://saifmohammad.com/WebDocs/AIT-2018/AIT2018-DATA/SemEval2018-Task1-all-data.zip
• SemEval 2022 - Task 6 (iSarcasmEval) : Intended Sarcasm Detection In English and Arabic - https://github.com/iabufarha/iSarcasmEval
• Annotating Offensive Language Datasets with Annotators’ Disagreement - https://github.com/dhfbk/annotators-agreement-dataset
• Large Scale Crowdsourcing and Characterization of Twitter Abusive Behavior - https://zenodo.org/record/3706866
• Learning With Disagreements (Le-Wi-Di), 2nd edition - https://codalab.lisn.upsaclay.fr/competitions/6146
• EmoThreat: Emotions & Threat Detection in Urdu - https://codalab.lisn.upsaclay.fr/competitions/5718 - https://sites.google.com/view/multi-label-emotionsfire-task/home
• TempoWiC - Detecting Meaning Shift in Social Media - https://codalab.lisn.upsaclay.fr/competitions/5360
• UMSAB, the Unified Multilingual Sentiment Analysis Benchmark - https://github.com/cardiffnlp/xlm-t#3---umsab-the-unified-multilingual-sentiment-analysis-benchmark
• Shared Task on Threat, Aggression and Cyberbullying Identification (TRAC - 2022) - https://codalab.lisn.upsaclay.fr/competitions/4753#learn_the_details-overview
• TRAC 2022 Shared Task on Bias, Threat and Aggression Identification in Context (BiTAg-Con) - https://codalab.lisn.upsaclay.fr/competitions/5792
• SMM4H 2022 - Task 1 - ADE Mining in English Tweets - https://codalab.lisn.upsaclay.fr/competitions/2073
• SMM4H 2022 - Task 9 Classification of social media forum posts self-reporting exact age - https://codalab.lisn.upsaclay.fr/competitions/3646
• Fine-Grained Hate Speech Detection on Arabic Twitter - https://codalab.lisn.upsaclay.fr/competitions/2324 https://codalab.lisn.upsaclay.fr/competitions/2332 https://codalab.lisn.upsaclay.fr/competitions/2334
• NaijaSenti: A Nigerian Twitter Sentiment Corpus for Multilingual Sentiment Analysis - https://zenodo.org/record/6538055
• CHIME dataset: Identifying and Categorizing Disaster-Related Tweets - https://github.com/Project-EPIC/chime-annotation
• A Survey and Experiments on Annotated Corpora for Emotion Classification in Text - https://github.com/sarnthil/unify-emotion-datasets/tree/master/datasets
• Stance Sentiment Emotion Corpus (SSEC): An annotation of the SemEval 2016 Twitter stance and sentiment corpus with emotion labels - https://www.ims.uni-stuttgart.de/forschung/ressourcen/korpora/stanceemotion/
• Hate Speech / Offensive Speech in the US 2020 Elections: Corpus for hate speech detection and stance detection - https://www.ims.uni-stuttgart.de/forschung/ressourcen/korpora/stance-hof/
• Visual Emotion Corpus: Visual Emotion Corpus - https://www.ims.uni-stuttgart.de/forschung/ressourcen/korpora/visual-emotion-corpus/
• DISARM: Detecting the Victims Targeted by Harmful Memes - https://github.com/Shiv681991/DISARM
• Multimodal-Sarcasm-Explanation--MuSE - https://github.com/LCS2-IIITD/Multimodal-Sarcasm-Explanation-MuSE
• Multilingual code-mixed datasets for classification - https://github.com/LCS2-IIITD/Code-mixed-classification/tree/main/data Same version
• Hinglish code-mixed hate detection - https://github.com/LCS2-IIITD/Hinglish_offense_detection-Neurocomputing2021
• Hindi Ho​st​ile Posts in ​Regional L​anguages dur​ing Emerge​ncy Si​tuation - https://competitions.codalab.org/competitions/26654#learn_the_details-dataset
• HSpam14 Dataset (14M tweets tagged for spam or not using various techniques) - https://personal.ntu.edu.sg/axsun/datasets.html
• Automatic Misogyny Identification (AMI) - https://amievalita2020.github.io/data/
• 2nd Hate Speech Detection Task at Evalita 2020 - https://github.com/msang/haspeede
• IronITA (Irony Detection in Italian Tweets) - http://www.di.unito.it/~tutreeb/ironita-evalita18
• Offensive language dataset of Croatian, English and Slovenian comments FRENK 1.1 - https://www.clarin.si/repository/xmlui/handle/11356/1462
• Extraction of Informative COVID-19 Danish Tweets - https://github.com/beaol/Finding-the-needle-in-a-haystack-Extraction-of-Informative-COVID-19-Danish-Tweets
• Tweet Topic Single Label - https://huggingface.co/datasets/cardiffnlp/tweet_topic_single
• Tweet Topic Multi Label - https://huggingface.co/datasets/cardiffnlp/tweet_topic_multi
• AfriSenti-SemEval Shared Task 12 - https://afrisenti-semeval.github.io/#starter-kit
• Customer Support on Twitter - https://www.kaggle.com/datasets/thoughtvector/customer-support-on-twitter
• NusaX - Multilingual parallel corpus for sentiment analysis https://github.com/IndoNLP/nusax
• CrisisFACTS is an open data challenge for state-of-the-art temporal summarization technologies to support disaster-response managers' use of online data sources during crisis events - https://crisisfacts.github.io
• ReactionGIF is a unique, first-of-its-kind dataset of 30K tweets and their GIF reactions - https://github.com/bshmueli/ReactionGIF

Stance detection

• Stance in Replies and Quotes (SRQ): A New Dataset For Learning Stance in Twitter Conversations - https://zenodo.org/record/3609277#.XyNbovhKidY
• Roumors dataset - https://github.com/vahedq/rumors/tree/master/data
• Twitter Stance re-annotated with emotion - http://www.romanklinger.de/ssec/
• Twitter Titling Corpus (stance towards presidents) - https://heidata.uni-heidelberg.de/dataset.xhtml?persistentId=doi:10.11588/data/IOHXDF
• Stance and Gender Detection in Tweets on Catalan Independence@Ibereval 2017 - https://stel.ub.edu/Stance-IberEval2017/
• Danish political stance dataset - https://figshare.com/articles/dataset/Danish_political_stance_dataset/12382592
• Twitter data annotated with stance toward multipe targets - http://www.site.uottawa.ca/~diana/resources/
• SemEval 2016 Task 6: Detecting Stance in Tweets - http://alt.qcri.org/semeval2016/task6/
• Tweet stance detection - https://github.com/kennyjoseph/constance
• SardiStance at EVALITA 2020 - https://github.com/mirkolai/evalita-sardistance

Tagging

• Xlime - multi layered annotations in german, spanish and Italian - Sentiment, POS, NER - https://github.com/lrei/xlime_twitter_corpus - https://www.clarin.si/repository/xmlui/handle/11356/1078
• UD_Italian-PoSTWITA: Italian Tweet Universal Dependencies - https://github.com/UniversalDependencies/UD_Italian-PoSTWITA Another version with more details
• Ark POS tagging using user embedding as features - https://github.com/bmurali1994/socialnets_postagging
• Ark POS tagged data with original tweet JSON - https://github.com/brendano/ark-tweet-nlp/tree/master/data/twpos-data-v0.3-original-tweets
• Multiple twitter datasets from IIIT Delhi (includes details on Indian political candidates, code switched POS and NER) - http://precog.iiitd.edu.in/resources.html
• Italian Tweet Named Entity Linking - https://github.com/swapUniba/neel-it-twitter
• Italian Twitter datasets for NER, POS, Supersense tagging: http://www.evalita.it/
• NEEL Microposts challenge - http://microposts2016.seas.upenn.edu/
• Yodie corpora: https://gate.ac.uk/applications/yodie.html
• English web text - https://github.com/UniversalDependencies/UD_English-EWT
• Twitter treebank - https://github.com/Oneplus/Tweebank
• Twitter UD dependency parsing for african american dialect as well as normal english - http://slanglab.cs.umass.edu/TwitterAAE/
• Linguistic Code-Switching Evaluation Benchmark - https://ritual.uh.edu/lince/
• Aspect Term-Polarity Co-Extraction - https://github.com/ArrowLuo/DOER
• Tweets mentinioning drugs and medication (tagging task) - https://healthlanguageprocessing.org/pubs/journals/kusuri/
• Drug entity linking on Tweets - https://healthlanguageprocessing.org/sharedtask2/smm4h-sharedtask-2017/ (additional tweet datasets - https://healthlanguageprocessing.org/software-and-downloads/)
• Tweet, FB, and WhatsApp Code-Switched POS tagging - http://amitavadas.com/Code-Mixing.html
• Lowlands Cost-sensitive POS tagging data - https://bitbucket.org/lowlands/costsensitive-data/src/master/
• Entity Profiling in Microblog Posts (Aspect phrases and Opinion Targets) - http://nlp.uned.es/~damiano/datasets/entityProfiling_ORM_Twitter.html
• Twitter Domain Adapted POS dataset - https://github.com/guitaowufeng/TPANN/tree/master/data
• Clickbait spotting and spoling - https://webis.de/data/webis-clickbait-22.html https://webis.de/data/webis-clickbait-17.html https://webis.de/data/webis-clickbait-16.html
• POS, NER, Semantic Role for disaster events - https://github.com/Project-EPIC/epic-annotation
• BioClaim: Biomedical claims in tweets, BEAR: Biomedical Entities and Relations in Tweets, CoVERT: Fact-checking labels and evidence for tweets about Covid-19 - https://www.ims.uni-stuttgart.de/forschung/ressourcen/korpora/bioclaim/
• Serbian Twitter training corpus ReLDI-NormTagNER-sr 2.1 - https://www.clarin.si/repository/xmlui/handle/11356/1240
• CURT: Fact Check Claim Spans on Tweets - https://github.com/LCS2-IIITD/DABERTA-EMNLP-2022
• TweetNERD - End to End Entity Linking Benchmark for Tweets - https://zenodo.org/record/6617192
• TempoWiC - An Evaluation Benchmark for Detecting Meaning Shift in Social Media - https://github.com/cardiffnlp/TempoWiC

NER datasets

• WNUT 2016: Twitter NER - https://github.com/aritter/twitter_nlp/tree/master/data/annotated/wnut16
• WNUT 2017: Emerging Entities dataset - https://github.com/leondz/emerging_entities_17 Guidelines
• Italian - http://neel-it.github.io/
• German, Italian, Spanish - https://github.com/lrei/xlime_twitter_corpus
• Hindi English code-mixed - https://github.com/SilentFlame/Named-Entity-Recognition
• Densely annotated Wikipedia - https://github.com/klout/opendata/tree/master/wiki_annotation
• Spanish English and Arabic Egyptian - https://code-switching.github.io/2018/#shared-task-id
• SocialDisNER corpus: gold standard annotations for detection of disease mentions in Spanish tweets - https://zenodo.org/record/6773099
• ProfNER corpus: gold standard annotations for profession detection in Spanish COVID-19 tweets - https://zenodo.org/record/4563995
• Many code-switched - http://emnlp2014.org/workshops/CodeSwitch/call.html
• Croatian - https://www.clarin.si/repository/xmlui/handle/11356/1241
• Serbian - https://www.clarin.si/repository/xmlui/handle/11356/1240
• Turkish - https://github.com/dkucuk/Tweet-Dataset-NER-SD
• Named Entity Recognition on Turkish Tweets: http://optima.jrc.it/Resources/2014_JRC_Twitter_TR_NER-dataset.zip
• French - http://cap2017.imag.fr/competition.html
• Slovenian - https://www.clarin.si/repository/xmlui/handle/11356/1123
• Telugu - https://github.com/anikethjr/NER_Telugu
• German Italian and Spanish - https://github.com/lrei/xlime_twitter_corpus
• Hindi - http://au-kbc.org/nlp/ESM-FIRE2015/
• Temporally-Informed Analysis of Named Entity Recognition - https://zenodo.org/record/3899040#.XyNTV_hKidY - https://github.com/shrutirij/temporal-twitter-corpus
• Code Switched NER - https://code-switching.github.io/2018/#shared-task-id
• Bullying Traces Data Set - https://research.cs.wisc.edu/bullying/data.html
• Hindi English Codemixed NER - https://github.com/SilentFlame/Named-Entity-Recognition
• Cyberthreat detection - https://github.com/ndionysus/multitask-cyberthreat-detection - https://github.com/ndionysus/twitter-cyberthreat-detection
• Lowlands LREC 2014 corrected + new benchmarks - https://bitbucket.org/lowlands/release/src/master/LREC2014/twitter_ner/
• ParsTwiNER: Parsi Twitter NER - https://github.com/overfit-ir/parstwiner
• MultiModal NER - https://github.com/RiTUAL-UH/multimodal_NER
• DFKI MobIE Corpus (formerly "DAYSTREAM Corpus") - 3,232 German-language documents for NER - https://github.com/dfki-nlp/mobie
• TwiCS: Microblog Entity Mention Detection with Multi-pass Lightweight Computations - https://github.com/dalakada/TwiCSv2/tree/master/data
• Tweebank NER - https://github.com/social-machines/TweebankNLP
• Information Extractor for Conversational Systems in Indian Languages (IECSIL, NER + REL, Hindi, Tamil, Malayalam, Telugu and Kannada)- https://github.com/BarathiGanesh-HB/ARNEKT-IECSIL Password=arnekt@iecsil2018
• Hashtag Segmentation Dataset - https://github.com/prashantkodali/HashSet
• Sentiment Scopes for Entity-Level Sentiment Analysis [en, es] - https://github.com/leodotnet/sentimentscope
• MobIE: A German Dataset for Named Entity Recognition, Entity Linking and Relation Extraction in the Mobility Domain - https://github.com/dfki-nlp/mobie
• DFKI Product Corpus - 174 English web pages and social media posts annotated for product and company named entities, and the relation CompanyProvidesProduct: https://github.com/DFKI-NLP/product-corpus
• DFKI SmartData Corpus, a dataset of 2598 German-language documents which has been annotated with fine-grained geo-entities, such as streets, stops and routes, as well as standard named entity types - https://github.com/DFKI-NLP/smartdata-corpus
• SMM4H'2022, Task 10 - Detection of disease mentions in tweets–SocialDisNER - https://codalab.lisn.upsaclay.fr/competitions/3531
• METS-CoV: A Dataset of Medical Entity and Targeted Sentiment on COVID-19 Related Tweets - https://github.com/YLab-Open/METS-CoV
• TweetBank-NER - https://github.com/mit-ccc/TweebankNLP
• TweetNER7 - https://huggingface.co/datasets/tner/tweetner7
• TweetNERD - End to End Entity Linking Benchmark for Tweets - https://zenodo.org/record/6617192

Entity Linking

• 282 langauge entity linking data based on Wikipedia - http://nlp.cs.rpi.edu/wikiann/
• NEEL Microposts challenge - http://microposts2016.seas.upenn.edu/
• Yodie corpora: https://gate.ac.uk/applications/yodie.html
• https://github.com/juand-r/entity-recognition-datasets
• Sense disambiguation/Entity Linking - https://github.com/lunafeng/ELTDS
• Twitter At the Grammy's - Entity Linking and Clustering - https://bitbucket.org/mdredze/tgx
• Entity Clustering - https://bitbucket.org/noandrews/phyloinf
• NEEL v2 + IE + IR data from Microsoft - https://www.microsoft.com/en-us/download/details.aspx?id=52530
• Twitter NEED - https://github.com/badiehm/TwitterNEED
• WSDM 2012 - https://github.com/dice-group/gerbil/issues/46#issuecomment-257821253 - https://edgar.meij.pro/dataset-adding-semantics-microblog-posts/?utm_source=bit.ly&utm_medium=linked&utm_campaign=myblog
• Multimodal Entity Linking in Tweets - https://github.com/OA256864/MEL_Tweets
• MobIE: A German Dataset for Named Entity Recognition, Entity Linking and Relation Extraction in the Mobility Domain - https://github.com/dfki-nlp/mobie
• Twitter Multimodal Entity Linking using user mentions - https://github.com/seukgcode/MEL-GHMFC/tree/main/datasets/Twitter-MEL
• Italian tweets Named Entity Linking - https://github.com/evalita2016/data
• Entity Linking in Tweets for Book and Movie domain - https://github.com/sujanucsc/IEL-Twitter
• Implicit-Entity-Linking-in-Tweets-Resources-and-Dataset (fine-grained NER and implicit and explicit entities, 7.5K tweets) - https://github.com/HawreH/Implicit-Entity-Recognition-and-Linking-in-Tweets-Resources-and-Dataset/tree/master/Dataset
• Implicit Entity Linking in Tweets (600 tweets) - https://github.com/sujanucsc/IEL-Twitter
• TweetNERD - End to End Entity Linking Benchmark for Tweets - https://zenodo.org/record/6617192

Relation Extraction

• REDIT - A relation extraction module for Tint - https://github.com/dhfbk/redit

Machine Translation

• TwitterMT - http://komunitatea.elhuyar.eus/tweetmt/resources/#Downloads
• Parallel text extracted from twitter and Weibo - http://www.cs.cmu.edu/~lingwang/microtopia/#twitter

Paraphrase identification

• Paraphrase identification - https://github.com/cocoxu/SemEval-PIT2015
• URL base large scale paraphrase dataset - https://github.com/lanwuwei/Twitter-URL-Corpus
• Twitter Paraphrase Corpus - https://github.com/jpcorb20/bet-backtranslation-paraphrase-experiment/tree/master/datasets

Rumour detection

• Twitter 15 and Twitter 16 rumour detection in tweet trees - https://github.com/majingCUHK/Rumor_RvNN
• PHEME rumour detection data - https://figshare.com/articles/PHEME_dataset_of_rumours_and_non-rumours/4010619
• RumDetect combined with PHEME - https://github.com/majingCUHK/Rumor_GAN - combined data - https://www.dropbox.com/s/46r50ctrfa0ur1o/rumdect.zip?dl=0
• RumorEval 2019 data - https://figshare.com/articles/RumourEval_2019_data/8845580
• Augmented dataset of rumours and non-rumours for rumour detection - https://zenodo.org/record/3269768#.XyNe3fhKidY
• Tweet Check-Worthiness - https://sites.google.com/view/clef2020-checkthat/tasks/tasks-1-5-check-worthiness?authuser=0
• Profiling fake news spreaders on Twitter - https://pan.webis.de/clef20/pan20-web/author-profiling.html
• Twitter Claim Detection from Online Content - https://github.com/LCS2-IIITD/LESA-EACL-2021
• Twitter Death Hoaxes dataset - https://figshare.com/articles/dataset/Twitter_Death_Hoaxes_dataset/5688811

Fact Checking

• On the Role of Images for Analyzing Claims in Social Media - https://zenodo.org/record/4592249
• Covid 19 claim detection - https://github.com/LCS2-IIITD/LESA-EACL-2021
• Evaluation of Approaches for the Clickbait Challenge at SemEval 2023 - Clickbait Spoiling - https://github.com/pan-webis-de/pan-code/tree/master/semeval23

Treebank and parsing

• English web text - https://github.com/UniversalDependencies/UD_English-EWT
• Twitter treebank - https://github.com/Oneplus/Tweebank
• Twitter UD dependency parsing for african american dialect as well as normal english - http://slanglab.cs.umass.edu/TwitterAAE/

Question answering

• Tweet QA - https://tweetqa.github.io/

Conversations

• Twitter Chat Corpus - https://github.com/marsan-ma/chat_corpus
• Microsoft Research Social Media Conversation Corpus - https://www.microsoft.com/en-us/download/details.aspx?id=52375
• Coreference Resolution in Twitter Conversations - https://github.com/verosol/e2e-coref-to-Twitter

Information Retrieval

• Signal-1M - https://research.signal-ai.com/datasets/signal1m-tweetir.html tweets data - Related to https://research.signal-ai.com/newsir16/signal-dataset.html

Multimodal

• Hate speech images - https://github.com/imatge-upc/hate-speech-detection
• Hate Speech Detection in Multimodal Publications - https://gombru.github.io/2019/10/09/MMHS/
• Text-Image Relationship in Twitter - https://github.com/danielpreotiuc/text-image-relationship
• Reaction GIF - Tweet and their GIF Reactions labeled with sarcarsm - https://github.com/bshmueli/ReactionGIF
• Target-Oriented Multimodal Sentiment Classification - https://github.com/jefferyYu/TomBERT/tree/master/absa_data

Sentence Similarity

• Twitter4SSE (Co-QuoteTweet, Co-Reply) - https://github.com/marco-digio/Twitter4SSE

Summarization

• Summarizing Tweets - https://github.com/cocoxu/twittersummarization
• Disaster Tweet Summarization - https://github.com/krudra/disaster_summarizer_TWEB_2018
• RepLab 2013 (Entity-Oriented Summaries for Reputation Management) - http://nlp.uned.es/replab2013/ https://zenodo.org/record/2536801#.XVrin-hKh3g
• ISSumSet: A Tweet Summarization Dataset Hidden in a TREC Track - https://github.com/AlexisDusart/ISSumSet
• Capitalizing on a TREC Track to Build a Tweet Summarization Dataset - https://github.com/AlexisDusart/SetSummTweet
• TES 2012-2016 tweet summary using Wikipedia Current Event portal - https://github.com/AlexisDusart/TSSuBERT/tree/main/TES%202012-2016
• Twitter event datasets (2012-2016, 30 events) - https://figshare.com/articles/dataset/Twitter_event_datasets_2012-2016_/5100460

Bot Detection

• TwiBot-22: Towards Graph-Based Twitter Bot Detection- https://github.com/LuoUndergradXJTU/TwiBot-22

RecSys

• Cross-Domain Rating Datasets from Structured Data on Twitter - https://github.com/sidooms/Twitter-ratings
• A Movie Rating Dataset Collected From Twitter - https://github.com/sidooms/MovieTweetings
• Twitter RecSys Challenge 2020 - http://www.recsyschallenge.com/2020/
• Twitter RecSys Challenge 2021 - https://www.recsyschallenge.com/2021/

Multi-Task

• SuperTweetEval: a unified benchmark of 12 heterogeneous NLP tasks - https://huggingface.co/datasets/cardiffnlp/super_tweeteval

General

• SocialLink is a publicly-available Linked Open Data dataset that matches social media accounts on Twitter to the corresponding entities in multiple language chapters of DBpedia - https://springernature.figshare.com/articles/dataset/ - GithubSocialLink_knowledge_transfer_between_social_media_and_linked_open_data/5235823/1
• Tweets informing about resource needs and availabilities in post-disaster situation - https://zenodo.org/record/2649794#.X7vpaGhKh3g
• Paraphrase identification - https://github.com/cocoxu/SemEval-PIT2015
• URL base large scale paraphrase dataset - https://github.com/lanwuwei/Twitter-URL-Corpus
• Language identification code-switched - http://www.care4lang.seas.gwu.edu/cs2/call.html
• Linked Open Data dataset that matches social media accounts on Twitter to the corresponding entities in multiple language chapters of DBpedia - https://zenodo.org/record/1451797#.XyNYM_hKidY
• Tweets in Space: Geo-tagged Tweet IDs 08/03/18 - 01/12/20 - https://zenodo.org/record/3608252#.XyNacfhKidY
• Fake Health News detection, which includes news contents, news reviews, social engagements and user network - https://zenodo.org/record/3862989#.XyNas_hKidY
• Tweets with reverse geo-tagging - https://zenodo.org/record/11661#.XyNfv_hKidY
• ULSN: A dataset for User Profile Linkage (From Twitter to Quora) - https://zenodo.org/record/3837711#.XyNgd_hKidY
• Multiple corpus for Twitter LangId, NER, etc - https://github.com/dimazest/2018-langid/tree/master/supplement
• Celebrity Profiling - Celebrity Tweet Profiles liked with demographic attributes and Wikidata - https://github.com/webis-de/ACL-19/tree/master/celebrity-profiling
• Topical Attention Election and Brexit - https://github.com/somethingx01/TopicalAttentionElection https://github.com/somethingx01/TopicalAttentionBrexit
• Hashtag segmentation - https://github.com/mounicam/hashtag_master
• English Twitter corpus (Geo-Tweets2019) built for training dialect-sensitive word embeddings - https://yuxingch.github.io/DialectGram/
• Many twitter datasets from ICWSM 2017 - https://zenodo.org/search?page=1&size=20&q=ICWSM
• Arabic English parallel Tweets - https://alt.qcri.org/resources/bilingual_corpus_of_parallel_tweets
• Voter Fraud Tweet dataset - https://voterfraud2020.io/
• Multiple Arabic Tweet datasets - http://qufaculty.qu.edu.qa/telsayed/datasets/
• Sentiment Analysis for Indian Languages (HI, BN, TM tweets) - http://amitavadas.com/SAIL/data.html (Download from: http://amitavadas.com/SAIL/Data/)
• Twitter customer care conversations dataset - https://github.com/IBM/twitter-customer-care-document-prediction
• WebLogo - 2m logo images from Twitter belonging to o specific logo types - https://weblogo2m.github.io/
• Disaster Tweet Corpus - https://zenodo.org/record/3713920
• Realtime classification of Twitter Trends - http://nlp.uned.es/~damiano/datasets/TT-classification.html
• RepLab 2014 (tweeet topic and tweet author reputation) - http://nlp.uned.es/replab2014/#dataset
• RepLab 2013 (entity relevance, polarity, clustering) - http://www.evall.uned.es/evaluate_using_benchmark#benchmark/description
• Papers and codes with Twitter datasets - https://index.quantumstat.com/#twitter
• Hateful user dataset - https://www.kaggle.com/manoelribeiro/hateful-users-on-twitter Source
• Dataset of public interest interventions on Twitter for politicians and candidates during the 2020 US General Election - https://github.com/SMAPPNYU/twitter_elections_public_interest
• A directory of local news outlets by state and medium. Includes social media accounts for local news outlets - https://github.com/yinleon/LocalNewsDataset
• Tweets about the Top Companies from 2015 to 2020 - https://www.kaggle.com/omermetinn/tweets-about-the-top-companies-from-2015-to-2020
• 40k full Twitter user profile data (including who they follow!) - https://www.kaggle.com/hwassner/TwitterFriends
• Twitter User Demographics Prediction - https://www.kaggle.com/crowdflower/twitter-user-gender-classification
• Twitter Social Network [ASU] - https://www.kaggle.com/mathurinache/twitter-edge-nodes Original Data Other OSN network data
• Twitter event datasets (2012-2016, 30 events) - https://figshare.com/articles/dataset/Twitter_event_datasets_2012-2016_/5100460
• SMILE Twitter Emotion - https://figshare.com/articles/dataset/smile_annotations_final_csv/3187909
• Early Risk prediction from social media - https://erisk.irlab.org/
• MediaEval 2020: Various datasets for media evaluation includes Fake News tweet data - https://multimediaeval.github.io/editions/2020/
• Smappdragon is a set of tools for working with twitter data - https://github.com/SMAPPNYU/smappdragon https://github.com/SMAPPNYU/pysmap
• MultiLexNorm: A Shared Task on Multilingual Lexical Normalization - https://bitbucket.org/robvanderg/multilexnorm/src/master/
• Ukraine and Russia Conflict Tweet IDs - https://github.com/echen102/ukraine-russia
• NELA-GT-2021: A Large Multi-Labelled News Dataset for The Study of Misinformation in News Articles (contains tweets embedded in news articles) - https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/RBKVBM
• Information Seeking in Academic Conferences - https://zenodo.org/record/819537
• TimeLMs: Diachronic Language Models Evaluation for Tweets - https://github.com/cardiffnlp/timelms
• EvoNLP Shared Task: Temporal Meaning Shift - https://sites.google.com/view/evonlp/shared-task
• Interactive Tagging Networks (Following/Followers and Tags on 1 million Twitter Users) - https://zenodo.org/record/16267
• User Movement Analysis - https://github.com/Project-EPIC/Twitter-Movement-Derivation/tree/master/All-Coded-Data

Tools, Tips, and Tricks

• General tricks on using the Twitter search API properly with custom filters - https://github.com/igorbrigadir/twitter-advanced-search
• Important time stamps in history of Twitter - https://github.com/igorbrigadir/twitter-history
• TweetedAt timestampt from tweet ID - https://github.com/oduwsdl/tweetedat
• Twitter API v2 tutorial - https://github.com/twitterdev/getting-started-with-the-twitter-api-v2-for-academic-research
• Twarc - https://github.com/DocNow/twarc/
• Facepager - https://github.com/strohne/Facepager
• Official Twitter Bulk Dataset downloader for Academic Use cases - https://developer.twitter.com/apitools/downloader
• Processing Tweets - https://colab.research.google.com/drive/1Uoog-scrKhLX3QZmrs1ClxCFdGIMH88i?usp=sharing#scrollTo=PmPc5qiIW6Fa
• Ekphrasis is a text processing tool - https://github.com/cbaziotis/ekphrasis
• SocialMediaIE - https://github.com/socialmediaie/SocialMediaIE
• TwitterNER - https://github.com/napsternxg/TwitterNER
• TweetNLP - https://tweetnlp.org/
• StoryWrangler: ngrams trends in Tweets - https://storywrangling.org/about Code
• Contagiograms: language specific retweet frequency - https://gitlab.com/compstorylab/contagiograms
• Happy Words from Hedonometer - https://hedonometer.org/words/labMT-en-v2/
• Twitter-Text Official Library for Parsing Tweets - https://github.com/twitter/twitter-text
• Twitter-Stanza - https://github.com/mit-ccc/TweebankNLP
• DMRC Academic Twitter Archive Collector - https://github.com/qut-dmrc/DMRC_Academic_Twitter_Archive_Collector

Embeddings

• Word Embedding Data Sets Learned from 400M Tweets and General Data - https://zenodo.org/record/581402#.XyNhUPhKidY
• Glove Word Embeddings - https://zenodo.org/record/3237458#.YEo-eZ1Kh3g
• Monthly word embeddings for Twitter random sample (English, 2012-2018) - https://zenodo.org/record/3527983
• Twitter Sent2Vec - https://github.com/epfml/sent2vec##downloading-pre-trained-models
• TimeLMs: Diachronic Language Models from Twitter - https://github.com/cardiffnlp/timelms
• Cross-lingual word embeddings from Twitter - https://github.com/pedrada88/crossembeddings-twitter
• Tweet Word Embeddings - https://tweetnlp.org/resources/

Unlabled topic specific data dumps

• Covid-19 Twitter chatter dataset for scientific use - http://www.panacealab.org/covid19/ https://zenodo.org/record/3960911#.XyNW9vhKidY
• Coronavirus Twitter Data: A collection of COVID-19 tweets with automated annotations - https://zenodo.org/record/3897727#.XyNY__hKidY
• Full trump tweet archive - https://github.com/bpb27/trump_tweet_data_archive
• Many US politicians tweet archive (periodically updated) - https://github.com/bpb27/political_twitter_archive
• TweetsCOV19 is a semantically annotated corpus of Tweets about the COVID-19 pandemic (8.1M) - https://data.gesis.org/tweetscov19/
• TweetsKB is a public RDF corpus of anonymized data for a large collection of annotated tweets (2B, periodically updated). - https://data.gesis.org/tweetskb/
• Twitter historical dataset: March 21, 2006 (first tweet) to July 31, 2009 (3 years, 1.5 billion tweets) - https://zenodo.org/record/3833782#.XyNRIvhKidY
• 307'061 verified Twitter accounts and their tweets from 2018-11-20 to 2018-11-28 - https://zenodo.org/record/2223647#.XyNTCvhKidY
• GeoCoV19: A Dataset of 524 Million of Multilingual COVID-19 Tweets with Location Information - https://zenodo.org/record/3878599

Facebook

Classification

• Facebook post valence and arousal - http://www.preotiuc.ro/resources.html
• FB What (or Who) Is Public? Privacy Settings and Social Media Content Sharing - https://github.com/compsocial/WhatWhoCSCW2017
• FB English Hindi Aggresion Identification - https://sites.google.com/view/trac1/shared-task
• FB News Page and Posts - https://bigquery.cloud.google.com/dataset/jbencina-144002:fb_news and https://github.com/jbencina/facebook-news/
• MultiEmotion-It - Italian Multiemotion corpus - https://github.com/RacheleSprugnoli/Esercitazioni_SA/tree/master/dataset
• Sinhala FB Decade Corpora https://github.com/LIRNEasia/FacebookDecadeCorpora
• Saved You a Click dataset - https://github.com/janetzhong/Saved-You-A-Click-CS224N/tree/main/data

General

• Personal Timeline tool - https://github.com/facebookresearch/personal-timeline

Instagram

Multimodal Tasks

• DankMemes 2020 - https://github.com/gianlucalebani/dankmemes2020

Youtube:

Classification

• Sentube sentiment - https://github.com/Abdelghafour01/sentube-sentiment
• Multimodal sentiment from video - https://github.com/soujanyaporia/multimodal-sentiment-analysis
• Tamil Malayalam Code-Mix Sentiment - https://dravidian-codemix.github.io/2020/datasets.html
• MultiEmotion-It - Italian Multiemotion corpus - https://github.com/RacheleSprugnoli/Esercitazioni_SA/tree/master/dataset
• JTubeSpeech: Corpus of speech collected from YouTube - https://github.com/sarulab-speech/jtubespeech
• Sentnob: Bengali Youtube Comments Sentiment Dataset - https://www.kaggle.com/cryptexcode/sentnob-sentiment-analysis-in-noisy-bangla-texts
• English YouTube Hate Speech Corpus - https://www.clarin.si/repository/xmlui/handle/11356/1454
• Italian YouTube Hate Speech Corpus - https://www.clarin.si/repository/xmlui/handle/11356/1450
• Sentiment Analysis and Homophobia detection of YouTube comments - https://codalab.lisn.upsaclay.fr/competitions/5310

Videos

• ClipShots: large-scale dataset for shot boundary detection collected from Youtube and Weibo covering more than 20 categories - https://github.com/Tangshitao/ClipShots

General

• Trending YouTube Video Statistics [multi-country] - https://www.kaggle.com/datasnaek/youtube-new
• SponsorBlock - SponsorBlock is an open-source crowdsourced browser extension and open API for skipping sponsor segments in YouTube videos - Database

Reddit

Classification

• Sarcasm detection - https://github.com/SenticNet/CASCADE
• Self-annotated sarcasm corpora - http://nlp.cs.princeton.edu/SARC/2.0/ Paper
• Conversational data - https://github.com/PolyAI-LDN/conversational-datasets
• GoEmotions: A Dataset of Fine-Grained Emotions - https://github.com/google-research/google-research/tree/master/goemotions
• Reddit Wallstreetbets Support and Intent Annotated Dataset 2021 - https://zenodo.org/record/5851847#%23.YeO_vhPML8E
• CL-Aff Shared Task: Get it #OffMyChest affective understanding of conversations - https://github.com/kj2013/claff-offmychest
• Us vs. Them: A Dataset of Populist Attitudes, News Bias and Emotions - https://github.com/LittlePea13/UsVsThem
• CAD: the Contextual Abuse Dataset - https://zenodo.org/record/4881008
• MMEmo Corpus: Multi-Modal Emotion Recognition Corpus of Reddit Posts - https://www.ims.uni-stuttgart.de/forschung/ressourcen/korpora/mmemo/
• Reddit News Chatter Intensity - https://github.com/LCS2-IIITD/ChatterNet
• DEBAGREEMENT: A comment-reply dataset for (dis)agreement detection in online debates - https://scale.com/open-av-datasets/oxford
• Author Topic Style Embeddings - https://github.com/nlpsoc/Style-Embeddings
• UnifiedQA Reddit SYAC (Saved you a click) - https://huggingface.co/marksverdhei/unifiedqa-large-reddit-syac - https://github.com/marksverdhei/reddit-syac

Summarization

• Reddit TIFU: Abstractive Summarization of Reddit Posts (120k) - https://github.com/ctr4si/MMN HF dataset

Sequence Tagging

• PASTRIE - MWE annotations - https://github.com/nert-nlp/pastrie

Named Entitis

• Entity Linking dataset - https://zenodo.org/record/3970806
• GUMReddit (Entity Linking, NER, etc) - https://github.com/universalDependencies/UD_English-GUMReddit/

Conversations

• Fast Adapation of Predicted User Responses in Goal-Oriented Dialogue - https://competitions.codalab.org/competitions/20152 Data Source HF data
• Reddit Writing Prompts - https://huggingface.co/datasets/rewardsignal/reddit_writing_prompts

Tools

• Data Tools for Reddit - https://github.com/dewarim/data-tools-for-reddit

Gab

• The Gab Hate Corpus - https://osf.io/edua3/
• Hateful user dataset - https://zenodo.org/record/5140191#.YQBGhI4zY2w Source
• Abuse Detection, Severity and Target Prediction for Gab Posts - https://github.com/mohit3011/AbuseAnalyzer

Summarization

• TL;DR corpus - https://zenodo.org/record/1168855#.X1WcDGdKidY

Amazon

• Irony and sarcasm in reviews - https://github.com/ef2020/SarcasmAmazonReviewsCorpus/wiki

eCommerce website in Italy

• Aspect Term Extraction and Aspect-based Sentiment Analysis Task - http://www.di.uniba.it/~swap/ate_absita/dataset.html

About.me

• User identities from multiple social networks - https://github.com/WING-NUS/aboutme

Whatsapp

• ICWSM tutorial on collecting WhatsApp data - https://users.ics.aalto.fi/kiran/whatsapp-tutorial/
• Tools for WhatsApp data collection - https://gvrkiran.github.io/whatsapp-tutorial/
• WhatsApp group data: https://dataverse.mpi-sws.org/dataset.xhtml?persistentId=doi:10.5072/FK2/CS7U5P
• Cyberbullying among Italian students - https://github.com/dhfbk/WhatsApp-Dataset

Delicious

• SocialBM0311 is a large-scale social tagging/bookmarking dataset collected from Delicious.com - http://www.zubiaga.org/datasets/socialbm0311/

Ask.fm

• Ask.fm Cyberbullying Corpus - https://ritual.uh.edu/cyberbullying-corpus/

Flickr

Popularity prediction

• Social Media Prediction Challenge - http://smp-challenge.com/

Conversations

• Multiple conversation datasets - https://convokit.cornell.edu/
• Yahoo News Annotated Comments Corpus - https://github.com/cnap/ynacc
• Multi-Domain Wizard-of-Oz dataset (MultiWOZ), a fully-labeled collection of human-human written conversations spanning over multiple domains and topics - https://github.com/budzianowski/multiwoz

News Comments

• A Dataset of Journalists' Interactions with Their Readership: When Should Article Authors Reply to Reader Comments?

• https://github.com/julian-risch/CIKM2020/

• User Engagement in Online Discussions - https://github.com/julian-risch/ICWSM2020
• Top Comment or Flop Comment? Predicting and Explaining User Engagement in Online News Discussions - https://github.com/julian-risch/ICWSM2020
• SFU Opinion and Comments Corpus - https://github.com/sfu-discourse-lab/SOCC
• FRENK-MMC-RTV Slavic news comments moderated by platform - https://www.clarin.si/repository/xmlui/handle/11356/1201 Paper
• FRENK-STYRIA-24sata 1.0 Croatian news comments moderated by platform - https://www.clarin.si/repository/xmlui/handle/11356/1202 Paper

Weibo

• Weibo-Douban user alignment across networks - https://github.com/ChenBaiyang/MAUIL
• Continuously updated Sina Weibo Public Opinion Datasets (only for research) - https://github.com/nghuyong/weibo-public-opinion-datasets
• Tencent Weibo (following network) - https://github.com/dedekinds/Graph-Embedding
• Weibo User Depression Detection Dataset - https://github.com/aidenwang9867/Weibo-User-Depession-Detection-Dataset
• ClipShots: large-scale dataset for shot boundary detection collected from Youtube and Weibo covering more than 20 categories - https://github.com/Tangshitao/ClipShots
• Multimodal entity linking (MEL) - https://github.com/seukgcode/MELBench
• Multilingual EN, JA, ZH Covid Weibo data - https://github.com/sociocom/covid19_dataset
• Social Media Keyphrase Generation - https://github.com/yuewang-cuhk/TAKG
• MELBench - Multimudal Entity Linking - https://github.com/seukgcode/MELBench
• Russia Ukraine Crisis Weibo (RUW) Dataset - https://github.com/yrf1/RussiaUkraine_weibo_dataset

TikTok

• MVIndEmo: A Dataset for Micro-video Public Induced Emotion Prediction on Social Media - https://github.com/inspur-hsslab/NeurIPS-Dataset-Induced-Emotion
• Pyktok: A simple module to collect video, text, and metadata from TikTok - https://github.com/dfreelon/pyktok

Whisper

• Whisper dataset - https://github.com/Mainack/whisper-2014-2016-data-HT-2020
• Hate Speech dataset - https://github.com/Mainack/hatespeech-data-HT-2017

SMS

• SMS Noun Phrase Detection - https://github.com/statnlp-research/statnlp-datasets/blob/master/dataset/SMSNP_data.zip

Stormfront

• Hate speech dataset - https://github.com/Vicomtech/hate-speech-dataset

Meneame

• Dataset of discussion threads from Meneame (Spanish) - https://zenodo.org/record/2536218

ShareChat

• Multilingual Abusive Comment Detection at Scale for Indic Languages - https://github.com/ShareChatAI/MACD
• Multilingual and Multimodal Abuse Detection - https://github.com/ShareChatAI/MADA
• 3MASSIV: Multilingual, Multimodal and Multi-Aspect dataset of Social Media Short Videos - https://github.com/ShareChatAI/3MASSIV
• ADIMA: ABUSE DETECTION IN MULTILINGUAL AUDIO - https://github.com/ShareChatAI/ADIMA

Koo App

• The Koo Dataset: An Indian Microblogging Platform With Global Ambitions (400m+ user interactions and 1.4m user profiles) - https://zenodo.org/records/10476212

General

• Pheme project: https://www.pheme.eu/software-downloads/
• Many pheme related datasets: https://figshare.com/authors/Arkaitz_Zubiaga/1376367
• Hatespeech term database: https://hatebase.org
• Hatespeech lexicons: http://hatespeech.di.unito.it/resources.html
• https://github.com/sebastianruder/NLP-progress
• 282 langauge entity linking data based on Wikipedia - http://nlp.cs.rpi.edu/wikiann/
• https://github.com/juand-r/entity-recognition-datasets
• Hashtag generation corpus - https://github.com/yuewang-cuhk/HashtagGeneration
• Klout FB and Twitter user reaction timestamps - https://github.com/klout/opendata/blob/master/when_to_post/README.md
• GSCL Shared Task: Automatic Linguistic Annotation of Computer-Mediated Communication / Social Media (POS and tokenization) https://sites.google.com/site/empirist2015/
• Matching identities across social media platforms - http://lig-membres.imag.fr/gogao/data.html
• Internet Argument Corpus v2 - https://nlds.soe.ucsc.edu/iac2
• Argument Facet Similarity Corpus - https://nlds.soe.ucsc.edu/node/44
• GDELT Social Media In the News - https://blog.gdeltproject.org/daily-updates-to-master-list-of-social-media-in-the-news-in-csv-and-bigquery/ and https://blog.gdeltproject.org/compiling-a-master-list-of-social-media-in-the-news-2016-2019/
• WKWSCI Sentiment Lexicon v1.1 - https://researchdata.ntu.edu.sg/dataset.xhtml?persistentId=doi:10.21979/N9/DWWEBV
• Voat.co dataset - https://zenodo.org/record/5841668
• The MeLa BitChute Dataset - https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/KRD1VS
• Effect of Popularity Shocks on User Behavior - https://zenodo.org/record/5854980#.YjgJAi8RoTs
• SemEval-2022 Shared Task 10: Structured Sentiment Analysis - https://github.com/jerbarnes/semeval22_structured_sentiment
• Dissecting harmful memes for Semantic role labelling of entities - https://codalab.lisn.upsaclay.fr/competitions/906#learn_the_details-evaluation
• Resources for Emotion Analysis: A collection of ressources created at IMS related to emotion and sentiment analysis - https://www.ims.uni-stuttgart.de/forschung/ressourcen/korpora/emotion/
• MHA-Meme-Leveraging-Sentence-Demarcations-and-Multi-hop-Attention-for-Meme-Affect-Analysis - https://github.com/LCS2-IIITD/MHA-MEME
• Misinfo text datasets - https://github.com/sfu-discourse-lab/MisInfoText
• Wikipedia Talk: A collection of data sets on Wikipedia Talk page discussions - https://figshare.com/projects/Wikipedia_Talk/16731
• The Misinformation Game: a social-media simulator - https://misinfogame.com/