A Parler (alternative social network) dataset of ~183,000,000 posts from Parler's full history before deplatforming, collected by Aliapoulios et al. via Parler API (2021). Each record contains post body text, username, timestamp, hashtags, upvotes, reposts, impressions, comment count, sensitive flag, media attachments, reply depth, parent post ID. Complete Parler post archive from before the platform's deplatforming — includes Jan 6 Capitol riot content, far-right discourse, and cross-posts from Twitter. Suitable for deplatforming research, far-right discourse analysis, January 6th research, hate speech detection, alternative platform behavior, content moderation effects.
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 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 twitter dataset of 767,139 tweet IDs from #BlackLivesMatter and #AllLivesMatter (10% Gardenhose sample, Aug 2014–Aug 2015), collected by Gallagher (Northeastern University, 2018). Each record contains tweet IDs only — full content requires hydration. Paired collection of #BLM and #ALM tweets enabling contrastive social movement analysis. Suitable for social movement analysis, racial discourse, political polarization.
A Twitter (user-hashtag interactions) dataset of 705,632 nodes and 1,890,661 edges, collected by Munmun De Choudhury et al. via Twitter API. Each record contains directed: False, weighted: True, signed: False, temporal: True. Bipartite temporal network linking Twitter users to hashtags they used, with Unix timestamps. Useful for studying hashtag adoption, information diffusion, and user interest modeling.. Suitable for hashtag analysis, bipartite network analysis, information diffusion.
A Bluesky (2024 presidential debate coverage) dataset of 279,249 bluesky posts from June 25 - July 2, 2024, collected by Zenodo contributors via public API. Each record contains post text, user ID, post ID, creation date, hashtags, mentioned users, URLs, post type (post/reply/repost), linked post, root post, language. 279,249 bluesky posts covering the 2024 Biden-Trump presidential debate with pre-computed sentence embeddings. Suitable for political communication, debate analysis, cross-platform comparison, decentralized social media research, NLP with embeddings.
A Twitter dataset of 6,184 tweets from Fortune 100 companies, collected by Kevin McElwee via Twitter API. Each record contains tweet text, corporation name, sector, timestamp, hashtags, racial justice label, BLM mention, Juneteenth mention, donation mention. A manually annotated dataset of tweets from Fortune 100 companies during the 2020 Black Lives Matter protests, capturing corporate responses to a major social movement. Suitable for corporate activism, social movement response, political communication, crisis communication, public relations, brand activism.
A Instagram (user engagement patterns) dataset of 1,000 posts with engagement metrics and multimedia, collected by Hanmante via Instagram API (2024). Each record contains post captions, hashtags, images, videos, likes, comments, user metadata, and tagged users. The dataset is notable for its dataset of Instagram political content from Uruguay, with rich multimodal data. Intended for research into social media engagement analysis, political communication, multimodal NLP, and user behavior modeling.
A #petpeeve (annoying behaviors) dataset of 3,375 tweets labeled with annoying behavior categories, collected by Wang and Yang via Twitter API (2015). Each record contains tweet text and annoying behavior category label. The dataset is notable for its dataset for automatic categorization of annoying behaviors using lexical and frame-semantic embeddings. Intended for research into social media analysis, text classification, and behavioral categorization.
Parler
Truth Social
TikTok
Twitter (X)
Bluesky
Instagram