A ReCOVery (news credibility repository) dataset of 2,029 news articles with reliability labels and social media engagement, collected by Zhou, Mulay, Ferrara, and Zafarani via news websites and Twitter (2020). Each record contains news article title and body, reliability label (1=reliable, 0=unreliable), publisher, political bias, country, author, and social media engagement data. The dataset is notable for its multimodal repository combining news content with social media engagement, designed for COVID-19 news credibility research. Intended for research into fake news detection, credibility assessment, political bias analysis, and multimodal misinformation research.
A Twitter dataset of 76,484 tweets across 19 events, 10 humanitarian categories, collected by Alam et al. for 19 natural disaster events (2016-2019). Each record contains tweet text, humanitarian category label, event name, train/dev/test split. Largest publicly available human-annotated crisis dataset with 76,484 tweets across 19 events. Suitable for crisis informatics, humanitarian response, text classification, social media analysis.
A Twitter dataset of 60,082 tweets across 6 crisis events (0 on-topic, 60,082 off-topic), collected by Olteanu et al. via crowdsourcing for 6 crisis events. Each record contains tweet text, on-topic/off-topic label (crowdsourced majority voting), crisis type, country, year. The dataset is notable for its Systematically collected dataset of 6 crisis events with crowdsourced on-topic/off-topic labels from at least 3 workers per tweet, enabling comparative analysis of tweet relevance across different crises. Cumulative version combines all events into a single corpus.. Intended for research into crisis informatics, tweet filtering, on-topic detection, information extraction, crowdsourcing evaluation, event detection, situation awareness, multi-event NLP benchmarking.
Twitter conversations around the Charlie Hebdo shooting in Paris (January 2015), from the PHEME rumour dataset (Zubiaga, Liakata, Procter, Kochkina and colleagues; the complete release for rumour detection and veracity classification, figshare 6392078): 2,079 threads, each a source tweet and its replies, of which 458 were judged rumours by journalists and 1,621 not. Every rumour also carries its veracity as later established: 193 true, 116 false and 149 unverified. 38,268 tweets in all, with full tweet JSON. ECHO builds the reply network (36,189 reply links, each reply pointing to the tweet it answers) and labels each tweet with its thread's rumour status and, for source tweets, the veracity. Useful for rumour detection, rumour verification and modelling how conversations unfold. Annotations CC BY 4.0; Twitter retains the rights to tweet content. The archive PHEME_veracity.tar.bz2 is gzip-compressed despite its name (the name is the authors' own, kept so the checksum matches figshare): extract it with `tar xzf`.
A Twitter dataset of 27,933 tweets across 26 crisis events (2012-2013), collected by Olteanu et al. via Twitter API during 26 crisis events. Each record contains tweet text, crisis type, country, event name. The dataset is notable for its Systematically collected dataset of 26 crisis events with balanced sampling across event types, enabling comparative analysis of social media communications across different crises. Cumulative version combines all events into a single corpus.. Intended for research into crisis informatics, disaster response, social media analysis, cross-crisis comparison, emergency communication patterns, situation awareness, multi-event NLP benchmarking.
Twitter conversations around the Ferguson unrest after the shooting of Michael Brown (August 2014), from the PHEME rumour dataset (Zubiaga, Liakata, Procter, Kochkina and colleagues; the complete release for rumour detection and veracity classification, figshare 6392078): 1,143 threads, each a source tweet and its replies, of which 284 were judged rumours by journalists and 859 not. Every rumour also carries its veracity as later established: 10 true, 8 false and 266 unverified. 24,175 tweets in all, with full tweet JSON. ECHO builds the reply network (22,260 reply links, each reply pointing to the tweet it answers) and labels each tweet with its thread's rumour status and, for source tweets, the veracity. Useful for rumour detection, rumour verification and modelling how conversations unfold. Annotations CC BY 4.0; Twitter retains the rights to tweet content. The archive PHEME_veracity.tar.bz2 is gzip-compressed despite its name (the name is the authors' own, kept so the checksum matches figshare): extract it with `tar xzf`.
Twitter conversations around the Sydney Lindt café siege (December 2014), from the PHEME rumour dataset (Zubiaga, Liakata, Procter, Kochkina and colleagues; the complete release for rumour detection and veracity classification, figshare 6392078): 1,221 threads, each a source tweet and its replies, of which 522 were judged rumours by journalists and 699 not. Every rumour also carries its veracity as later established: 382 true, 86 false and 54 unverified. 23,996 tweets in all, with full tweet JSON. ECHO builds the reply network (22,775 reply links, each reply pointing to the tweet it answers) and labels each tweet with its thread's rumour status and, for source tweets, the veracity. Useful for rumour detection, rumour verification and modelling how conversations unfold. Annotations CC BY 4.0; Twitter retains the rights to tweet content. The archive PHEME_veracity.tar.bz2 is gzip-compressed despite its name (the name is the authors' own, kept so the checksum matches figshare): extract it with `tar xzf`.
A Twitter dataset of 16,057 tweets across 7 crisis events referencing 17,465 images by URL, collected by Alam et al. for 7 major natural disasters in 2017. Each record contains tweet text, engagement metrics (retweets, favorites, replies, quotes), user metadata, media attachments, crisis code. The dataset is notable for its Systematically collected multimodal dataset of 7 natural disasters with full Twitter API metadata. Cumulative version combines all events into a single corpus. Intended for research into crisis informatics, multimodal learning, disaster response, image classification, tweet filtering, cross-event engagement analysis. This copy (the SocialED redistribution) holds the tweet objects only: the images are referenced by URL and the image files are not included, and neither are CrisisMMD's human annotations (informativeness, humanitarian category, damage severity).
Twitter conversations around the Ottawa Parliament Hill shooting (October 2014), from the PHEME rumour dataset (Zubiaga, Liakata, Procter, Kochkina and colleagues; the complete release for rumour detection and veracity classification, figshare 6392078): 890 threads, each a source tweet and its replies, of which 470 were judged rumours by journalists and 420 not. Every rumour also carries its veracity as later established: 329 true, 72 false and 69 unverified. 12,284 tweets in all, with full tweet JSON. ECHO builds the reply network (11,394 reply links, each reply pointing to the tweet it answers) and labels each tweet with its thread's rumour status and, for source tweets, the veracity. Useful for rumour detection, rumour verification and modelling how conversations unfold. Annotations CC BY 4.0; Twitter retains the rights to tweet content. The archive PHEME_veracity.tar.bz2 is gzip-compressed despite its name (the name is the authors' own, kept so the checksum matches figshare): extract it with `tar xzf`.
A Twitter dataset of 10,033 tweets (0 on-topic, 10,033 off-topic), collected by Olteanu et al. via crowdsourcing for the Queensland Floods 2013 crisis. Each record contains tweet text, on-topic/off-topic label (crowdsourced majority voting), crisis type, country, year. The dataset is notable for its tweets from the Queensland Floods 2013 crisis, labeled by crowdsourcing workers for on-topic/off-topic classification with majority voting from at least 3 workers per tweet. Intended for research into crisis informatics, tweet filtering, on-topic detection, information extraction, crowdsourcing evaluation, event detection, situation awareness.
A Twitter dataset of 10,031 tweets (0 on-topic, 10,031 off-topic), collected by Olteanu et al. via crowdsourcing for the Alberta Floods 2013 crisis. Each record contains tweet text, on-topic/off-topic label (crowdsourced majority voting), crisis type, country, year. The dataset is notable for its tweets from the Alberta Floods 2013 crisis, labeled by crowdsourcing workers for on-topic/off-topic classification with majority voting from at least 3 workers per tweet. Intended for research into crisis informatics, tweet filtering, on-topic detection, information extraction, crowdsourcing evaluation, event detection, situation awareness.
A Twitter dataset of 10,012 tweets (0 on-topic, 10,012 off-topic), collected by Olteanu et al. via crowdsourcing for the Boston Bombings 2013 crisis. Each record contains tweet text, on-topic/off-topic label (crowdsourced majority voting), crisis type, country, year. The dataset is notable for its tweets from the Boston Bombings 2013 crisis, labeled by crowdsourcing workers for on-topic/off-topic classification with majority voting from at least 3 workers per tweet. Intended for research into crisis informatics, tweet filtering, on-topic detection, information extraction, crowdsourcing evaluation, event detection, situation awareness.
A Twitter dataset of 10,008 tweets (0 on-topic, 10,008 off-topic), collected by Olteanu et al. via crowdsourcing for the Sandy Hurricane 2012 crisis. Each record contains tweet text, on-topic/off-topic label (crowdsourced majority voting), crisis type, country, year. The dataset is notable for its tweets from the Sandy Hurricane 2012 crisis, labeled by crowdsourcing workers for on-topic/off-topic classification with majority voting from at least 3 workers per tweet. Intended for research into crisis informatics, tweet filtering, on-topic detection, information extraction, crowdsourcing evaluation, event detection, situation awareness.
A Twitter dataset of 10,006 tweets (0 on-topic, 10,006 off-topic), collected by Olteanu et al. via crowdsourcing for the West Texas Explosion 2013 crisis. Each record contains tweet text, on-topic/off-topic label (crowdsourced majority voting), crisis type, country, year. The dataset is notable for its tweets from the West Texas Explosion 2013 crisis, labeled by crowdsourcing workers for on-topic/off-topic classification with majority voting from at least 3 workers per tweet. Intended for research into crisis informatics, tweet filtering, on-topic detection, information extraction, crowdsourcing evaluation, event detection, situation awareness.
A Twitter dataset of 9,992 tweets (0 on-topic, 9,992 off-topic), collected by Olteanu et al. via crowdsourcing for the Oklahoma Tornado Season 2013 crisis. Each record contains tweet text, on-topic/off-topic label (crowdsourced majority voting), crisis type, country, year. The dataset is notable for its tweets from the Oklahoma Tornado Season 2013 crisis, labeled by crowdsourcing workers for on-topic/off-topic classification with majority voting from at least 3 workers per tweet. Intended for research into crisis informatics, tweet filtering, on-topic detection, information extraction, crowdsourcing evaluation, event detection, situation awareness.
A Twitter dataset of 9,399 tweets from hurricane_irma_2017, collected by Alam et al. for hurricane_irma_2017 crisis. Each record contains tweet text, humanitarian category label, train/dev/test split. Human-annotated crisis tweets from the hurricane_irma_2017 event. Suitable for crisis informatics, humanitarian response, text classification.
A Twitter dataset of 9,112 tweets from hurricane_harvey_2017, collected by Alam et al. for hurricane_harvey_2017 crisis. Each record contains tweet text, humanitarian category label, train/dev/test split. Human-annotated crisis tweets from the hurricane_harvey_2017 event. Suitable for crisis informatics, humanitarian response, text classification.
A Twitter dataset of 7,984 tweets from kerala_floods_2018, collected by Alam et al. for kerala_floods_2018 crisis. Each record contains tweet text, humanitarian category label, train/dev/test split. Human-annotated crisis tweets from the kerala_floods_2018 event. Suitable for crisis informatics, humanitarian response, text classification.
A Twitter dataset of 7,613 tweets from hurricane_dorian_2019, collected by Alam et al. for hurricane_dorian_2019 crisis. Each record contains tweet text, humanitarian category label, train/dev/test split. Human-annotated crisis tweets from the hurricane_dorian_2019 event. Suitable for crisis informatics, humanitarian response, text classification.
A Twitter dataset of 7,376 tweets from california_wildfires_2018, collected by Alam et al. for california_wildfires_2018 crisis. Each record contains tweet text, humanitarian category label, train/dev/test split. Human-annotated crisis tweets from the california_wildfires_2018 event. Suitable for crisis informatics, humanitarian response, text classification.
A Twitter dataset of 7,278 tweets from hurricane_maria_2017, collected by Alam et al. for hurricane_maria_2017 crisis. Each record contains tweet text, humanitarian category label, train/dev/test split. Human-annotated crisis tweets from the hurricane_maria_2017 event. Suitable for crisis informatics, humanitarian response, text classification.
A Twitter dataset of 6,264 tweets from hurricane_florence_2018, collected by Alam et al. for hurricane_florence_2018 crisis. Each record contains tweet text, humanitarian category label, train/dev/test split. Human-annotated crisis tweets from the hurricane_florence_2018 event. Suitable for crisis informatics, humanitarian response, text classification.
A Twitter dataset of 4,952 tweets (1,238 rumour, 3,714 non-rumour) from the 2013 Boston Marathon bombings, collected by Zubiaga, Liakata, and Procter — Augmented PHEME (Zenodo, 2019). Each record contains tweet text, binary rumour/non-rumour label, timestamp, reaction thread structure linking source tweets to replies. Augmented extension of PHEME covering the Boston Marathon bombings event with both rumour threads and non-rumour threads, providing a balanced benchmark for rumour detection with reaction graph structure. Suitable for rumour detection, misinformation classification, crisis communication analysis, stance detection, conversational thread modelling.
Twitter conversations around the Germanwings plane crash in the French Alps (March 2015), from the PHEME rumour dataset (Zubiaga, Liakata, Procter, Kochkina and colleagues; the complete release for rumour detection and veracity classification, figshare 6392078): 469 threads, each a source tweet and its replies, of which 238 were judged rumours by journalists and 231 not. Every rumour also carries its veracity as later established: 94 true, 111 false and 33 unverified. 4,489 tweets in all, with full tweet JSON. ECHO builds the reply network (4,020 reply links, each reply pointing to the tweet it answers) and labels each tweet with its thread's rumour status and, for source tweets, the veracity. Useful for rumour detection, rumour verification and modelling how conversations unfold. Annotations CC BY 4.0; Twitter retains the rights to tweet content. The archive PHEME_veracity.tar.bz2 is gzip-compressed despite its name (the name is the authors' own, kept so the checksum matches figshare): extract it with `tar xzf`.
A Twitter dataset of 4,021 tweets referencing 3,611 images by URL, collected by Alam et al. for the Hurricane Irma 2017 crisis. Each record contains tweet text, engagement metrics (retweets, favorites, replies, quotes), user metadata, media attachments, crisis code. The dataset is notable for its multimodal dataset from the Hurricane Irma 2017 crisis, collected from Twitter during the event with full Twitter API metadata. Intended for research into crisis informatics, multimodal learning, disaster response, image classification, tweet filtering, engagement analysis. This copy (the SocialED redistribution) holds the tweet objects only: the images are referenced by URL and the image files are not included, and neither are CrisisMMD's human annotations (informativeness, humanitarian category, damage severity).
A Twitter dataset of 3,995 tweets referencing 4,691 images by URL, collected by Alam et al. for the Hurricane Maria 2017 crisis. Each record contains tweet text, engagement metrics (retweets, favorites, replies, quotes), user metadata, media attachments, crisis code. The dataset is notable for its multimodal dataset from the Hurricane Maria 2017 crisis, collected from Twitter during the event with full Twitter API metadata. Intended for research into crisis informatics, multimodal learning, disaster response, image classification, tweet filtering, engagement analysis. This copy (the SocialED redistribution) holds the tweet objects only: the images are referenced by URL and the image files are not included, and neither are CrisisMMD's human annotations (informativeness, humanitarian category, damage severity).
A Twitter dataset of 3,992 tweets referencing 4,569 images by URL, collected by Alam et al. for the Hurricane Harvey 2017 crisis. Each record contains tweet text, engagement metrics (retweets, favorites, replies, quotes), user metadata, media attachments, crisis code. The dataset is notable for its multimodal dataset from the Hurricane Harvey 2017 crisis, collected from Twitter during the event with full Twitter API metadata. Intended for research into crisis informatics, multimodal learning, disaster response, image classification, tweet filtering, engagement analysis. This copy (the SocialED redistribution) holds the tweet objects only: the images are referenced by URL and the image files are not included, and neither are CrisisMMD's human annotations (informativeness, humanitarian category, damage severity).
A Twitter dataset of 3,933 tweets from cyclone_idai_2019, collected by Alam et al. for cyclone_idai_2019 crisis. Each record contains tweet text, humanitarian category label, train/dev/test split. Human-annotated crisis tweets from the cyclone_idai_2019 event. Suitable for crisis informatics, humanitarian response, text classification.
A Twitter dataset of 2,242 tweets from canada_wildfires_2016, collected by Alam et al. for canada_wildfires_2016 crisis. Each record contains tweet text, humanitarian category label, train/dev/test split. Human-annotated crisis tweets from the canada_wildfires_2016 event. Suitable for crisis informatics, humanitarian response, text classification.
A Twitter dataset of 2,195 tweets from kaikoura_earthquake_2016, collected by Alam et al. for kaikoura_earthquake_2016 crisis. Each record contains tweet text, humanitarian category label, train/dev/test split. Human-annotated crisis tweets from the kaikoura_earthquake_2016 event. Suitable for crisis informatics, humanitarian response, text classification.
A Twitter dataset of 2,015 tweets from puebla_mexico_earthquake_2017, collected by Alam et al. for puebla_mexico_earthquake_2017 crisis. Each record contains tweet text, humanitarian category label, train/dev/test split. Human-annotated crisis tweets from the puebla_mexico_earthquake_2017 event. Suitable for crisis informatics, humanitarian response, text classification.
A Twitter dataset of 1,958 tweets from pakistan_earthquake_2019, collected by Alam et al. for pakistan_earthquake_2019 crisis. Each record contains tweet text, humanitarian category label, train/dev/test split. Human-annotated crisis tweets from the pakistan_earthquake_2019 event. Suitable for crisis informatics, humanitarian response, text classification.
A Twitter dataset of 1,880 tweets from midwestern_us_floods_2019, collected by Alam et al. for midwestern_us_floods_2019 crisis. Each record contains tweet text, humanitarian category label, train/dev/test split. Human-annotated crisis tweets from the midwestern_us_floods_2019 event. Suitable for crisis informatics, humanitarian response, text classification.
A Twitter dataset of 1,654 tweets from hurricane_matthew_2016, collected by Alam et al. for hurricane_matthew_2016 crisis. Each record contains tweet text, humanitarian category label, train/dev/test split. Human-annotated crisis tweets from the hurricane_matthew_2016 event. Suitable for crisis informatics, humanitarian response, text classification.
A Twitter dataset of 1,563 tweets from ecuador_earthquake_2016, collected by Alam et al. for ecuador_earthquake_2016 crisis. Each record contains tweet text, humanitarian category label, train/dev/test split. Human-annotated crisis tweets from the ecuador_earthquake_2016 event. Suitable for crisis informatics, humanitarian response, text classification.
A Twitter dataset of 1,515 tweets from greece_wildfires_2018, collected by Alam et al. for greece_wildfires_2018 crisis. Each record contains tweet text, humanitarian category label, train/dev/test split. Human-annotated crisis tweets from the greece_wildfires_2018 event. Suitable for crisis informatics, humanitarian response, text classification.
A Twitter dataset of 1,486 tweets referencing 1,590 images by URL, collected by Alam et al. for the California Wildfires 2017 crisis. Each record contains tweet text, engagement metrics (retweets, favorites, replies, quotes), user metadata, media attachments, crisis code. The dataset is notable for its multimodal dataset from the California Wildfires 2017 crisis, collected from Twitter during the event with full Twitter API metadata. Intended for research into crisis informatics, multimodal learning, disaster response, image classification, tweet filtering, engagement analysis. This copy (the SocialED redistribution) holds the tweet objects only: the images are referenced by URL and the image files are not included, and neither are CrisisMMD's human annotations (informativeness, humanitarian category, damage severity).
A Twitter dataset of 1,442 tweets, collected by Olteanu et al. via Twitter API during the 2013 Russia meteor crisis. Each record contains tweet text, crisis type, country, event name. The dataset is notable for its tweets from the 2013 Russia meteor crisis, collected systematically across 26 events for comparative analysis. Intended for research into crisis informatics, disaster response, social media analysis, cross-crisis comparison, emergency communication patterns, situation awareness.
A Twitter dataset of 1,412 tweets, collected by Olteanu et al. via Twitter API during the 2012 Costa Rica earthquake crisis. Each record contains tweet text, crisis type, country, event name. The dataset is notable for its tweets from the 2012 Costa Rica earthquake crisis, collected systematically across 26 events for comparative analysis. Intended for research into crisis informatics, disaster response, social media analysis, cross-crisis comparison, emergency communication patterns, situation awareness.
A Twitter dataset of 1,250 tweets, collected by Olteanu et al. via Twitter API during the 2013 Savar building collapse crisis. Each record contains tweet text, crisis type, country, event name. The dataset is notable for its tweets from the 2013 Savar building collapse crisis, collected systematically across 26 events for comparative analysis. Intended for research into crisis informatics, disaster response, social media analysis, cross-crisis comparison, emergency communication patterns, situation awareness.
A Twitter dataset of 1,237 tweets referencing 1,380 images by URL, collected by Alam et al. for the Mexico Earthquake 2017 crisis. Each record contains tweet text, engagement metrics (retweets, favorites, replies, quotes), user metadata, media attachments, crisis code. The dataset is notable for its multimodal dataset from the Mexico Earthquake 2017 crisis, collected from Twitter during the event with full Twitter API metadata. Intended for research into crisis informatics, multimodal learning, disaster response, image classification, tweet filtering, engagement analysis. This copy (the SocialED redistribution) holds the tweet objects only: the images are referenced by URL and the image files are not included, and neither are CrisisMMD's human annotations (informativeness, humanitarian category, damage severity).
A Twitter dataset of 1,201 tweets from italy_earthquake_aug_2016, collected by Alam et al. for italy_earthquake_aug_2016 crisis. Each record contains tweet text, humanitarian category label, train/dev/test split. Human-annotated crisis tweets from the italy_earthquake_aug_2016 event. Suitable for crisis informatics, humanitarian response, text classification.
A Twitter dataset of 1,200 tweets, collected by Olteanu et al. via Twitter API during the 2012 Colorado wildfires crisis. Each record contains tweet text, crisis type, country, event name. The dataset is notable for its tweets from the 2012 Colorado wildfires crisis, collected systematically across 26 events for comparative analysis. Intended for research into crisis informatics, disaster response, social media analysis, cross-crisis comparison, emergency communication patterns, situation awareness.
A Twitter dataset of 1,200 tweets, collected by Olteanu et al. via Twitter API during the 2013 Queensland floods crisis. Each record contains tweet text, crisis type, country, event name. The dataset is notable for its tweets from the 2013 Queensland floods crisis, collected systematically across 26 events for comparative analysis. Intended for research into crisis informatics, disaster response, social media analysis, cross-crisis comparison, emergency communication patterns, situation awareness.
A Twitter dataset of 1,199 tweets, collected by Olteanu et al. via Twitter API during the 2013 Australia bushfire crisis. Each record contains tweet text, crisis type, country, event name. The dataset is notable for its tweets from the 2013 Australia bushfire crisis, collected systematically across 26 events for comparative analysis. Intended for research into crisis informatics, disaster response, social media analysis, cross-crisis comparison, emergency communication patterns, situation awareness.
A Twitter dataset of 1,100 tweets, collected by Olteanu et al. via Twitter API during the 2013 Glasgow helicopter crash crisis. Each record contains tweet text, crisis type, country, event name. The dataset is notable for its tweets from the 2013 Glasgow helicopter crash crisis, collected systematically across 26 events for comparative analysis. Intended for research into crisis informatics, disaster response, social media analysis, cross-crisis comparison, emergency communication patterns, situation awareness.
A Twitter dataset of 1,050 tweets, collected by Olteanu et al. via Twitter API during the 2012 Guatemala earthquake crisis. Each record contains tweet text, crisis type, country, event name. The dataset is notable for its tweets from the 2012 Guatemala earthquake crisis, collected systematically across 26 events for comparative analysis. Intended for research into crisis informatics, disaster response, social media analysis, cross-crisis comparison, emergency communication patterns, situation awareness.
A Twitter dataset of 1,048 tweets, collected by Olteanu et al. via Twitter API during the 2013 Typhoon Yolanda crisis. Each record contains tweet text, crisis type, country, event name. The dataset is notable for its tweets from the 2013 Typhoon Yolanda crisis, collected systematically across 26 events for comparative analysis. Intended for research into crisis informatics, disaster response, social media analysis, cross-crisis comparison, emergency communication patterns, situation awareness.
A Twitter dataset of 1,032 tweets, collected by Olteanu et al. via Twitter API during the 2013 LA airport shootings crisis. Each record contains tweet text, crisis type, country, event name. The dataset is notable for its tweets from the 2013 LA airport shootings crisis, collected systematically across 26 events for comparative analysis. Intended for research into crisis informatics, disaster response, social media analysis, cross-crisis comparison, emergency communication patterns, situation awareness.
A Twitter dataset of 1,000 tweets, collected by Olteanu et al. via Twitter API during the 2012 Italy earthquakes crisis. Each record contains tweet text, crisis type, country, event name. The dataset is notable for its tweets from the 2012 Italy earthquakes crisis, collected systematically across 26 events for comparative analysis. Intended for research into crisis informatics, disaster response, social media analysis, cross-crisis comparison, emergency communication patterns, situation awareness.
Twitter (X)