A TextDetox 2024 shared task (Russian) dataset of 5,000 Russian posts (2,500 toxic, 2,500 non-toxic), collected by Dementieva et al. for PAN/CLEF TextDetox 2024. Each record contains post text, binary toxicity label (0=non-toxic, 1=toxic). TextDetox Russian provides 5,000 balanced binary toxicity annotations — part of a 15-language multilingual benchmark for cross-lingual toxicity detection research. Suitable for toxicity detection, multilingual content moderation, cross-lingual NLP, hate speech classification, text detoxification.
A TextDetox 2024 shared task (Tatar) dataset of 5,000 Tatar posts (2,500 toxic, 2,500 non-toxic), collected by Dementieva et al. for PAN/CLEF TextDetox 2024. Each record contains post text, binary toxicity label (0=non-toxic, 1=toxic). TextDetox Tatar provides 5,000 balanced binary toxicity annotations — part of a 15-language multilingual benchmark for cross-lingual toxicity detection research. Suitable for toxicity detection, multilingual content moderation, cross-lingual NLP, hate speech classification, text detoxification.
A TextDetox 2024 shared task (Ukrainian) dataset of 5,000 Ukrainian posts (2,500 toxic, 2,500 non-toxic), collected by Dementieva et al. for PAN/CLEF TextDetox 2024. Each record contains post text, binary toxicity label (0=non-toxic, 1=toxic). TextDetox Ukrainian provides 5,000 balanced binary toxicity annotations — part of a 15-language multilingual benchmark for cross-lingual toxicity detection research. Suitable for toxicity detection, multilingual content moderation, cross-lingual NLP, hate speech classification, text detoxification.
A TextDetox 2024 shared task (Chinese) dataset of 5,000 Chinese posts (2,500 toxic, 2,500 non-toxic), collected by Dementieva et al. for PAN/CLEF TextDetox 2024. Each record contains post text, binary toxicity label (0=non-toxic, 1=toxic). TextDetox Chinese provides 5,000 balanced binary toxicity annotations — part of a 15-language multilingual benchmark for cross-lingual toxicity detection research. Suitable for toxicity detection, multilingual content moderation, cross-lingual NLP, hate speech classification, text detoxification.
A TheMarker Cafe dataset of 69,413 users and 1,644,849 undirected friendship links, collected by Fire et al. (2011) and distributed by Network Repository. TheMarker Cafe was an Israeli online community where most interaction was in Hebrew. Suitable for link prediction, social network analysis, community detection and regional online community studies.
Twitter History and Image Sharing (THIS) v1.0: country-level aggregates of about 14.6 billion geolocated tweets, 2013 to 2023, released by Zachary Steinert-Threlkeld on Harvard Dataverse. Four tables, 17,928,584 rows in all: tweets and users per country and day (792,313 rows), tweets per country, language and day (15,407,849), new accounts per country and day (943,361), and image-sharing availability per country and day (785,061). No individual tweets or users are included. Suitable for country-level trends in platform use, language use and image sharing.
A TikTok dataset of 1,035,817 videos, 4,920 creators, 27,169,858 engagement records, collected by Lingbo et al.. Each record contains video_id, author_id, create_time, description, hashtags, music, daily engagement (plays, likes, comments, shares, collects). Large-scale TikTok dataset with video metadata and 30-day engagement trajectories. Suitable for engagement analysis, creator behavior, viral prediction, social media dynamics.
A TikTok dataset of 209,543 videos from 1,872 creators, collected by Lingbo et al.. Each record contains video_id, author_id, create_time, description, transcript, GPT summary, topic, emotion scores, music metadata. Large-scale TikTok dataset with video metadata, GPT-generated summaries, emotion scores, and 30-day engagement trajectories. Suitable for engagement analysis, creator behavior, viral prediction, social media dynamics.
A Chicago regional road network (TNTP benchmark) dataset of 1,467 nodes and 1,298 edges, collected by Transportation Networks for Research (TNTP). Each record contains directed: False, weighted: False. Road network of the Chicago metropolitan area — classic transportation network benchmark.. Suitable for transportation network analysis, routing algorithms, network flow.
A Internet router topology dataset of 34,761 nodes and 107,720 edges, collected by KONECT via traceroute measurements. Each record contains directed: False, weighted: True. Temporal internet router topology network with Unix timestamps per measurement.. Suitable for internet topology analysis, network robustness, temporal network analysis.
A Web tracking ecosystem dataset of 40,421,974 nodes and 140,613,762 edges, collected by KONECT via web crawl. Each record contains directed: False, weighted: False. Network of web trackers co-occurring on the same websites — useful for studying the tracking ecosystem and privacy implications.. Suitable for web tracking analysis, privacy research, network structure of advertising.
A Twitter dataset of 4,465 labeled Twitter accounts (991 bots, 3474 genuine), collected by Ashu273k — curated subset of Cresci et al. (2017). Each record contains user profile features (followers, friends, statuses, favourites, listed count), binary bot/genuine label, user bio text. Small, clean subset of the Cresci-2017 dataset curated for rapid prototyping and benchmarking of bot detection models, with pre-extracted user-level features and binary labels. Suitable for bot detection, social spambot classification, feature selection for account authenticity, rapid prototyping of detection algorithms.
A snapshot of Truth Social from February to September 2022, released by Patrick Gerard, Nicholas Botzer and Tim Weninger (University of Notre Dame) on Zenodo (7522646). 454,458 users and 845,060 posts ('truths'), with 4,002,115 follow edges, 506,276 replies and 10,508 quotes, plus hashtags, media and external URLs, as 13 tab-separated tables. Suitable for social network analysis, political communication and misinformation research on an alternative platform.
A Truth Social (user account corpus) dataset of 90,683 user accounts with 1,611 verified and 0 bot-labeled, collected by TruthStance project team via Truth Social API. Each record contains username, follower/following counts, account creation date, profile bio text, bot label, verified status. TruthStance covers 90,683 Truth Social accounts spanning 2022-02 to 2025-12, with 1,611 verified accounts and 61,226 users with bio text — one of few datasets focused on Truth Social platform users. Suitable for Truth Social platform analysis, bot detection, political social media research, user behavior modeling, alternative platform studies.
A TUAW (The Unofficial Apple Weblog) dataset of 10,485 blog posts, collected by Agarwal, Tang, Liu, and Yu via TUAW crawl (2004-2007). Each record contains blog post title, text, author, categories, publication date, inlinks, outlinks, and comment counts. The dataset is notable for its blog dataset with rich metadata, representing a closed community of Apple-focused bloggers. Intended for research into influential blogger identification, blog post analysis, text classification, and influence modeling.
A Twitter (#BlackLivesMatter) dataset of 9,165 tweets (3,084 hate speech, 6,081 non-hate), collected by Giorgi et al. via Twitter API (2020). Each record contains tweet text, binary hate label (0/1), hashtags, source platform, user friends/favourites count, retweet flag. 9,165 BLM-related tweets with hate speech labels — captures online discourse during the 2020 Black Lives Matter movement with 3,084 hate speech examples. Suitable for hate speech detection, social movement analysis, BLM discourse research, Twitter content moderation, political communication, racial justice NLP.
A Twitch (gaming social network) dataset of 168,116 users and 6,797,558 mutual follower edges, collected by Rozemberczki & Sarkar via Twitch public API (2018). Each record contains undirected mutual follower edges, node features (views, mature, life_time, created_at, updated_at, dead_account, language, affiliate). The dataset is notable for its large-scale strongly connected component with no missing attributes, designed for node classification and count data regression. Intended for research into social network analysis, node classification, community detection, and transfer learning in gaming communities.
A Twitch dataset of 1,000 top streamers with viewership and engagement metrics, collected by Aayush Mishra via Twitch API (2020). Each record contains channel name, watch time, streaming time, max/average viewers, followers, followers gained, views gained, partnership status, age-restriction status, and language. A comprehensive snapshot of the top 1,000 Twitch streamers, capturing key engagement and growth metrics for the platform's most influential content creators. Suitable for influencer analysis, engagement prediction, platform dynamics, streamer growth modeling, gaming community studies.
A Twitter (2010 snapshot) dataset of 41,652,230 users and 1,468,365,182 directed follower edges, collected by Kwak, Lee, Park & Moon via Twitter API (2010). Each record contains source user ID, follower user ID (edge from i to j means j follows i). Complete snapshot of the Twitter follower graph from 2010 — one of the largest social network datasets ever released publicly, capturing the full follower structure of 41.6M Twitter users with 1.47 billion directed edges.. Suitable for large-scale social network analysis, community detection, influence propagation, graph algorithm benchmarking, Twitter network evolution studies.
A Twitter dataset of 11,929 labeled financial tweets, collected by zeroshot via Twitter API. Each record contains tweet text, sentiment label (positive/negative/neutral), sentiment value (0/1/2), train/valid split. A labeled dataset of tweets about financial news, with sentiment annotations for stock market-related content. Suitable for sentiment analysis, financial NLP, stock market prediction, social media analytics, natural language processing.
A Twitter dataset of 456,631 users, 14,855,875 follow/mention edges, collected by De Domenico et al. (2013) via Twitter API around Higgs boson announcement (July 4, 2012). Each record contains directed edges between Twitter users representing follows or mentions, capturing information propagation during a major scientific announcement. Temporal network capturing information spreading around the Higgs boson discovery announcement, enabling study of scientific rumor propagation and information cascades in real-time social media. Suitable for information propagation, rumor dynamics, temporal network analysis, scientific communication studies, social media epidemiology.
A Twitter (English corpus, 2B+ tweets) dataset of 11 years of monthly n-gram frequency counts (2013-2023), collected by Zenodo contributors via Twitter corpus analysis. Each record contains n-gram text, n-gram type (1=unigram, 2=bigram, etc.), monthly frequency count; both case-sensitive and lowercase variants. Monthly n-gram frequency distributions from over 2 billion English tweets spanning 2013-2023, enabling longitudinal linguistic trend analysis on Twitter at scale. Suitable for longitudinal linguistic analysis, trending topic detection, language change over time, Twitter vocabulary evolution, NLP corpus statistics, temporal word frequency research.
A Twitter dataset of 11,316,811 users and 85,331,843 directed follower/followee edges, collected by Zafarani and Liu at Syracuse University Data Lab. Each record contains directed edges from follower to followee; bare integer node IDs (anonymized users). one of the largest publicly available Twitter follower networks, with 85M directed edges enabling large-scale social network analysis at the full platform scale. Suitable for large-scale network analysis, influence propagation, community detection, centrality computation, and information diffusion modeling.
A Twitter (Cannes Film Festival 2013) dataset of 438,537 users, 991,854 multiplex edges (retweets, mentions, replies), collected by D. R. G. and D. M.. Each record contains source, target, weight, layer (1=retweet, 2=mention, 3=reply). Multiplex Twitter network during Cannes Film Festival 2013 with three interaction layers. Suitable for multiplex network analysis, event dynamics, information propagation.
A Twitter (MLK Day 2013) dataset of 327,707 users, 396,671 multiplex edges (retweets, mentions, replies), collected by D. R. G. and D. M.. Each record contains source, target, weight, layer (1=retweet, 2=mention, 3=reply). Multiplex Twitter network during MLK Day 2013 with three interaction layers. Suitable for multiplex network analysis, event dynamics, information propagation.
A Twitter (Moscow Athletics 2013) dataset of 88,804 users, 210,250 multiplex edges (retweets, mentions, replies), collected by D. R. G. and D. M.. Each record contains source, target, weight, layer (1=retweet, 2=mention, 3=reply). Multiplex Twitter network during Moscow Athletics 2013 with three interaction layers. Suitable for multiplex network analysis, event dynamics, information propagation.
A Twitter (NY Climate March 2014) dataset of 102,439 users, 353,495 multiplex edges (retweets, mentions, replies), collected by D. R. G. and D. M.. Each record contains source, target, weight, layer (1=retweet, 2=mention, 3=reply). Multiplex Twitter network during NY Climate March 2014 with three interaction layers. Suitable for multiplex network analysis, event dynamics, information propagation.
A Twitter (Obama in Israel 2013) dataset of 2,281,259 users, 4,061,960 multiplex edges (retweets, mentions, replies), collected by D. R. G. and D. M.. Each record contains source, target, weight, layer (1=retweet, 2=mention, 3=reply). Multiplex Twitter network during Obama in Israel 2013 with three interaction layers. Suitable for multiplex network analysis, event dynamics, information propagation.
The five Twitter multiplex event networks (Cannes 2013, MLK Day 2013, Moscow Athletics 2013, NY Climate March 2014, Obama in Israel 2013) side by side in one graph: 3,238,746 users and 5,175,900 weighted retweet, mention and reply edges. Each release numbers its users independently, so node IDs are event:user and no account is linked across events. Suitable for multiplex network analysis and comparing information spreading across events.
A Small group signed social network (balance theory study) dataset of 16 nodes and 58 edges, collected by UCI Network Data Repository. Each record contains directed: False, weighted: True, signed: True. Small signed social network with +1/-1 edges — useful for testing structural balance theory.. Suitable for signed network analysis, structural balance theory, community detection.
A University karate club (W.W. Zachary observation study) dataset of 34 nodes and 78 edges, collected by W.W. Zachary. Each record contains directed: False, weighted: False, signed: False. The most widely cited network in graph research — 34-node karate club that split into two factions, used to validate virtually every community detection algorithm.. Suitable for community detection benchmark, foundational social network analysis.
A Unicode CLDR language data dataset of 868 nodes and 1,255 edges, collected by KONECT via Unicode CLDR. Each record contains directed: False, weighted: True. Bipartite weighted network of languages and Unicode scripts, with edge weights representing similarity scores.. Suitable for linguistic network analysis, language typology, script similarity.
A Multiple (Wikipedia, GDELT, Google Trends) dataset of 51 weeks of cross-platform attention data across 6 regions, collected by Luke Steuber via Wikipedia API, GDELT Project, and Google Trends (2020-2025). Each record contains weekly attention scores by region, component metrics (Google Trends, GDELT, Wikipedia), and regional composite scores. A unified, weekly time-series of public attention to the United States from three independent sources, broken down by global region, enabling multi-perspective analysis of attention dynamics. Suitable for information cascade, event detection, public health monitoring, political communication, media analysis, attention tracking, regional sentiment analysis.
A Twitter/X dataset of 42,014,825 tweets from the 2024 US presidential election period across 39 data parts with full engagement metadata (41,561,521 records with engagement metrics), collected by USC Information Sciences Institute via Twitter/X API during the 2024 US election period. Each record contains tweet text, engagement metrics (reply count, retweet count, like count, quote count, view count), language, media attachments, reply chain metadata, user location, and publication timestamp. The dataset is notable for its one of the largest publicly available datasets from the 2024 US presidential election, with full engagement metadata per tweet enabling study of information amplification and political discourse dynamics. Intended for research into political opinion mining, election discourse analysis, misinformation detection, engagement prediction, and temporal analysis of political social media.
A Usenet (sample) dataset of 6,111 posts from 2470 newsgroups, collected by OwnedByDanes via Hugging Face. Each record contains post text, newsgroup, date, author, subject. Sample of Usenet posts from 1980–2013, with 5,000 posts per top‑level hierarchy. Covers comp.*, sci.*, rec.*, alt.*, etc.. Suitable for historical discourse analysis, early internet culture, topic evolution, authorship attribution.
A Usenet (sample) dataset of 7,361 posts from 801 newsgroups, collected by OwnedByDanes via Hugging Face. Each record contains post text, newsgroup, date, author, subject. Sample of Usenet posts from 1980–2013, with 5,000 posts per top‑level hierarchy. Covers comp.*, sci.*, rec.*, alt.*, etc.. Suitable for historical discourse analysis, early internet culture, topic evolution, authorship attribution.
A Usenet (sample) dataset of 7,361 posts from 7 newsgroups, collected by OwnedByDanes via Hugging Face. Each record contains post text, newsgroup, date, author, subject. Sample of Usenet posts from 1980–2013, with 5,000 posts per top‑level hierarchy. Covers comp.*, sci.*, rec.*, alt.*, etc.. Suitable for historical discourse analysis, early internet culture, topic evolution, authorship attribution.
A Usenet (sample) dataset of 7,361 posts from 182 newsgroups, collected by OwnedByDanes via Hugging Face. Each record contains post text, newsgroup, date, author, subject. Sample of Usenet posts from 1980–2013, with 5,000 posts per top‑level hierarchy. Covers comp.*, sci.*, rec.*, alt.*, etc.. Suitable for historical discourse analysis, early internet culture, topic evolution, authorship attribution.
A Usenet (sample) dataset of 7,361 posts from 53 newsgroups, collected by OwnedByDanes via Hugging Face. Each record contains post text, newsgroup, date, author, subject. Sample of Usenet posts from 1980–2013, with 5,000 posts per top‑level hierarchy. Covers comp.*, sci.*, rec.*, alt.*, etc.. Suitable for historical discourse analysis, early internet culture, topic evolution, authorship attribution.
A Usenet (sample) dataset of 7,361 posts from 710 newsgroups, collected by OwnedByDanes via Hugging Face. Each record contains post text, newsgroup, date, author, subject. Sample of Usenet posts from 1980–2013, with 5,000 posts per top‑level hierarchy. Covers comp.*, sci.*, rec.*, alt.*, etc.. Suitable for historical discourse analysis, early internet culture, topic evolution, authorship attribution.
A Usenet (sample) dataset of 64,999 posts from 4754 newsgroups, collected by OwnedByDanes via Hugging Face. Each record contains post text, newsgroup, date, author, subject. Sample of Usenet posts from 1980–2013, with 5,000 posts per top‑level hierarchy. Covers comp.*, sci.*, rec.*, alt.*, etc.. Suitable for historical discourse analysis, early internet culture, topic evolution, authorship attribution.
A Usenet (sample) dataset of 7,361 posts from 215 newsgroups, collected by OwnedByDanes via Hugging Face. Each record contains post text, newsgroup, date, author, subject. Sample of Usenet posts from 1980–2013, with 5,000 posts per top‑level hierarchy. Covers comp.*, sci.*, rec.*, alt.*, etc.. Suitable for historical discourse analysis, early internet culture, topic evolution, authorship attribution.
A Usenet (sample) dataset of 7,361 posts from 273 newsgroups, collected by OwnedByDanes via Hugging Face. Each record contains post text, newsgroup, date, author, subject. Sample of Usenet posts from 1980–2013, with 5,000 posts per top‑level hierarchy. Covers comp.*, sci.*, rec.*, alt.*, etc.. Suitable for historical discourse analysis, early internet culture, topic evolution, authorship attribution.
A Usenet (sample) dataset of 7,361 posts from 43 newsgroups, collected by OwnedByDanes via Hugging Face. Each record contains post text, newsgroup, date, author, subject. Sample of Usenet posts from 1980–2013, with 5,000 posts per top‑level hierarchy. Covers comp.*, sci.*, rec.*, alt.*, etc.. Suitable for historical discourse analysis, early internet culture, topic evolution, authorship attribution.
A Voat (Reddit alternative, 2013–2020) dataset of 3,201 submissions, 45,426 comments, 7,455 users, 20,751 collaboration edges, collected by Mekacher & Papasavva (ICWSM 2022) via full platform crawl. Each record contains user nodes (registration date, moderation status), directed weighted edges (commenter→submitter, weight=upvotes+1), submission titles and bodies, comment text, upvotes, downvotes, depth. Complete dataset for the Voat subverse v/4chan, covering its full lifetime on the platform (2013–2020). Includes all 3,201 submissions and 45,426 comments with full text and engagement metadata.. Suitable for social network analysis, community dynamics, content moderation, platform migration studies, online community research.
Posts from v/8chan on Voat, from the Voat archive of Mekacher & Papasavva (ICWSM 2022): 5,804 submissions and 630,059 comments with their text, votes and dates. The archive did not recover the author of almost any post in this subverse, so the entry has no user or reply graph. Suitable for studying anonymous posting, topic modelling and moderation.
Posts from v/anon on Voat, from the Voat archive of Mekacher & Papasavva (ICWSM 2022): 7,471 submissions and 32,011 comments with their text, votes and dates. The archive did not recover the author of almost any post in this subverse, so the entry has no user or reply graph. Suitable for studying anonymous posting, topic modelling and moderation.
Posts from v/AnonAll on Voat, from the Voat archive of Mekacher & Papasavva (ICWSM 2022): 34,658 submissions and 65,541 comments with their text, votes and dates. The archive did not recover the author of almost any post in this subverse, so the entry has no user or reply graph. Suitable for studying anonymous posting, topic modelling and moderation.
Posts from v/AnonTalk on Voat, from the Voat archive of Mekacher & Papasavva (ICWSM 2022): 10,459 submissions and 47,670 comments with their text, votes and dates. The archive did not recover the author of almost any post in this subverse, so the entry has no user or reply graph. Suitable for studying anonymous posting, topic modelling and moderation.
A Voat (Reddit alternative, 2013–2020) dataset of 2,956 submissions, 28,473 comments, 4,968 users, 17,262 collaboration edges, collected by Mekacher & Papasavva (ICWSM 2022) via full platform crawl. Each record contains user nodes (registration date, moderation status), directed weighted edges (commenter→submitter, weight=upvotes+1), submission titles and bodies, comment text, upvotes, downvotes, depth. Complete dataset for the Voat subverse v/askgoat, covering its full lifetime on the platform (2013–2020). Includes all 2,956 submissions and 28,473 comments with full text and engagement metadata.. Suitable for social network analysis, community dynamics, content moderation, platform migration studies, online community research.
TheMarker
Twitter (X)
TikTok
The internet
Truth Social
Twitch
Google
Wikipedia
Usenet
Voat