Content Of Tweets Research Articles

You Have to Send the Right Message: Examining the Influence of Protective Action Guidance on Message Perception Outcomes across Prior Hazard Warning Experience to Three Hazards

Abstract A long-term goal for warning-message designers is to determine the most effective type of message that can instruct individuals to act quickly and prevent loss of life and/or injury when faced with an imminent threat. One likely way to increase an individual’s behavioral intent to act when they are faced with risk information is to provide protective action information or guidance. This study investigated participant perceptions (understanding, believing, personalizing, deciding, milling, self-efficacy, and response efficacy) in response to the National Weather Service’s experimental product Twitter messages for three hazard types (tornado, snow squall, and dust storm), with each message varying by inclusion and presentation of protective action information placed in the tweet text and the visual graphic. We also examine the role of prior hazard warning experience on message perception outcomes. To examine the effects, the experiment used a between-subjects design in which participants were randomly assigned to one hazard type and received one of four warning messages. Participants then took a post-test measuring message perceptions, efficacy levels, prior hazard warning experience, and demographics. The results showed that, for each hazard and prior hazard experience level, messages with protective action guidance in both the text and graphic increase their understanding, belief, ability to decide, self-efficacy, and response efficacy. These results reinforce the idea that well-designed messages that include protective action guidance work well regardless of hazard type or hazard warning experience. Significance Statement Preventing injury and/or loss of life during a hazardous event is a prime concern for disaster communicators. The study provides insights to practitioners on how to effectively communicate protective actions to audiences with varying familiarity with the hazard through Twitter posts. We experimented with tweet message design and content for three hazards: tornado, snow squall, and dust storm, to find that posts that include protective action guidance in both the text and image increase participant perceptions that they could perform the suggested protective actions, regardless of hazard type or hazard warning experience. Given our findings, practitioners should consider including protective action guidance in message text and graphic to warn members of the public with varied prior warning experience.

Geospatial Mapping of Suicide-Related Tweets and Sentiments among Malaysians during the COVID-19 Pandemic

The government enacted the Movement Control Order (MCO) to curb the spread of the COVID-19 pandemic in Malaysia, restricting movement and shutting down several commercial enterprises around the nation. The crisis, which lasted over two years and featured a few MCOs, had an impact on Malaysians’ mental health. This study aimed to understand the context of using the word “suicide” on Twitter among Malaysians during the pandemic. “Suicide” is a keyword searched for on Twitter when mining data with the NCapture plugin. Using NVivo 12 software, we used the content analysis approach to detect the theme of tweets discussed by tweeps. The tweet content was then analyzed using VADER sentiment analysis to determine if it was positive, negative, or neutral. We conducted a spatial pattern distribution of tweets, revealing high numbers from Kuala Lumpur, Klang, Subang Jaya, Kangar, Alor Setar, Chukai, Kuantan, Johor Bharu, and Kota Kinabalu. Our analysis of tweet content related to the word “suicide” revealed three (3) main themes: (i) criticism of the government of that day (CGD) (N = 218, 55.68%), (ii) awareness related to suicide (AS) (N = 162, 41.44%), and (iii) suicidal feeling or experience (SFE) (N = 12, 2.88%). The word “suicide” conveyed both negative and positive sentiments. Negative tweets expressed frustration and disappointment with the government’s response to the pandemic and its economic impact. In contrast, positive tweets spread hope, encouragement, and support for mental health and relationship building. This study highlights the potential of social-media big data to understand the users’ virtual behavior in an unprecedented pandemic situation and the importance of considering cultural differences and nuances in sentiment analysis. The spatial pattern information was useful in identifying areas that may require additional resources or interventions to address suicide risk. This study underscores the importance of timely and cost-effective social media data analysis for valuable insights into public opinion and attitudes toward specific topics.

THE BEHAVIORIAL PHENOMENON OF FRIEND WITH BENEFIT OF ALTER BASE @FWBSMG ON TWITTER

The emergence of alter ego accounts on Twitter triggers a new phenomenon called "friend with benefit," known as FWB. This phenomenon arises because there is a place to share information and communicate with each other. This FWB behavior is mushrooming on social media with an alter ego account, a second account aimed at seeking FWB, and a means of more unrestrained self-expression. This study aims to describe the FWB phenomenon, identify the forms of FWB behavior, and determine how the alter ego's motives account for FWB behavior. The method used in this research is qualitative content analysis with a qualitative approach. The method obtained is based on observations of @FWBSmg tweet content, interviews, and document review. The results of this study indicate that the FWB phenomenon carried out by alter ego accounts on Twitter is carried out because they want to find satisfaction both sexually and emotionally. Not only that but FWB is also conducted to find friends to chat and hang out with. The results of this study also show that the alter ego account is used as a second account to search for FWB and that the pattern used by the alter ego in searching for FWB on Twitter is also supported by the alter base community named @FWBSmg. Alter ego accounts send menfess with the gender terms "M" for male and "F" for a female when looking for FWB. They are also sending menfess photos by covering their faces with stickers; which is done to hide their real identities in public.

Sentiment Classification on Twitter Social Media Using K-Means Clustering, C4.5 and Naive Bayes (Case Study: Blocking Paypal by Kominfo)

Kominfo (Ministry of Communication and Information) requires all PSEs (Electronic System Providers) to register themselves so that their access is not blocked, as shown in the case of Paypal and several other PSEs. The blocking case reaps mixed opinions from netizens, especially Twitter social media users. We use the sentiment values obtained from the content of tweets collected through the crawling process and employ the K-Means Clustering to group them into clusters. Finally, we use these clusters as the target in a dataset and classify them using the C4.5 and Naive Bayes algorithms. Of the 1000 netizen tweets studied, we found that 6.5% of netizens supported the blocking action, 75.4% did not care or felt that the blocking action had no effect on them, and 15.4% did not support the blocking by Kominfo. The classification results in this study resulted in a 98.2% accuracy value, a 95% precision value, and a 95.5% recall value.

Dynamics of social media behavior before and after SARS-CoV-2 infection.

Online social media have been both a field of research and a source of data for research since the beginning of the COVID-19 pandemic. In this study, we aimed to determine how and whether the content of tweets by Twitter users reporting SARS-CoV-2 infections changed over time. We built a regular expression to detect users reporting being infected, and we applied several Natural Language Processing methods to assess the emotions, topics, and self-reports of symptoms present in the timelines of the users. Twelve thousand one hundred and twenty-one twitter users matched the regular expression and were considered in the study. We found that the proportions of health-related, symptom-containing, and emotionally non-neutral tweets increased after users had reported their SARS-CoV-2 infection on Twitter. Our results also show that the number of weeks accounting for the increased proportion of symptoms was consistent with the duration of the symptoms in clinically confirmed COVID-19 cases. Furthermore, we observed a high temporal correlation between self-reports of SARS-CoV-2 infection and officially reported cases of the disease in the largest English-speaking countries. This study confirms that automated methods can be used to find digital users publicly sharing information about their health status on social media, and that the associated data analysis may supplement clinical assessments made in the early phases of the spread of emerging diseases. Such automated methods may prove particularly useful for newly emerging health conditions that are not rapidly captured in the traditional health systems, such as the long term sequalae of SARS-CoV-2 infections.

Identify the Subject and Content of Tweets on Twitter Using Multilayer Neural Network Method and Random Graphs

Identify the Subject and Content of Tweets on Twitter Using Multilayer Neural Network Method and Random Graphs

Identifying Persian bots on Twitter; which feature is more important: Account Information or Tweet Contents?

Identifying Persian bots on Twitter; which feature is more important: Account Information or Tweet Contents?

A lexicon-based approach for sentiment analysis of multimodal content in tweets

A lexicon-based approach for sentiment analysis of multimodal content in tweets

Trends in Pro-Gun Control Tweets Surrounding a Mass Shooting Event Suggest Time-Dependent Changes in Public Response

Over the period from May 25, 2015 to May 25, 2018, there were 1,296 mass shootings recorded around the United States. This article evaluates overall public support for gun regulation before and after a major mass shooting. Big data that accounted for outliers and other potential skewing points were included from Twitter. 250 Tweets that occurred immediately before, immediately after, or two weeks after a major mass shooting were analyzed. The Tweets were restricted to include at least one of the following hashtags (#): secondamendment, neveragain, shooting, NRA, guncontrol, gunviolence, and 2A. The Tweets were categorized as pro-gun control, anti-gun control, or neutral, defined by operational definitions. It was hypothesized that there would be more Tweets that opposed gun control 1-7 days before each mass shooting but fewer Tweets that opposed gun control 72 hours after each mass shooting. Three days (72 hours) after a major mass shooting, there was a significant decrease in the number of anti-gun control Tweets posted compared to 1-7 days prior to the shooting. There was also a significant increase in the number of anti-gun control Tweets 14-21 days after the mass shooting compared to 72 hours after the shooting, implying that the content of Tweets after two weeks was similar to the content before the shooting occurred.

A two-year follow-up: Twitter activity regarding misinformation about spinal manipulation, chiropractic care and boosting immunity during the COVID-19 pandemic

BackgroundSpinal manipulative therapy (SMT) is offered by many health professions, most often by chiropractors. While SMT can be effective for some musculoskeletal disorders, there is no evidence that SMT improves human immunity in a clinically meaningful way. Despite this, we showed previously that Twitter misinformation about chiropractic/SMT improving immunity increased sharply at the start of the COVID-19 pandemic. Here, we perform a two-year follow-up.MethodsWe previously employed specialized software (i.e. Talkwalker) to search the entirety of Twitter activity in the months before and after the COVID-19 pandemic was declared (March 11, 2020). In this paper, we conducted follow-up searches over two successive 12 month periods using terms related to SMT, immunity and chiropractic. The resulting tweets were then coded into those promoting/refuting a relation between SMT and immunity (tone) and messaging about chiropractic/interventions (content). Further analyses were performed to subcategorize tweet content, tally likes, retweets and followers, and evaluate refuting tweets and the country of origin. Finally, we created a chronology of Twitter activity superimposed with dates of promoting or refuting activities undertaken by chiropractic organizations.ResultsOver the 27 month study period, Twitter activity peaked on March 31, 2020 then declined continuously. As in our first paper, our follow-up data showed that (1) the ratio of refuting/promoting tweets remained constant and (2) tweets that refuted a relationship between SMT and immunity were substantially more liked, retweeted and followed than those promoting. We also observed that promoting tweets suggesting that SMT improves immunity decreased more rapidly. Overwhelmingly, promoting tweets originated in the USA while refuting tweets originated in Canada, Europe and Australia. The timing of the decline in peak Twitter activity, together with a parallel decline in tweets claiming that SMT improves immunity, was coincident with initiatives by chiropractic organizations and regulators targeting misinformation.ConclusionOverwhelmingly, Twitter activity during the COVID-19 pandemic focussed on refuting a relation between chiropractic/SMT and immunity. A decline in Twitter activity promoting a relation between SMT and immunity was observed to coincide with initiatives from chiropractic organizations and regulators to refute these claims. The majority of misinformation about this topic is generated in the United States.

Responses to heat waves: what can Twitter data tell us?

Increasing average temperatures and heat waves are having devasting impacts on human health and well-being but studies of heat impacts and how people adapt are rare and often confined to specific locations. In this study, we explore how analysis of conversations on social media can be used to understand how people feel about heat waves and how they respond. We collected global Twitter data over four months (from January to April 2022) using predefined hashtags about heat waves. Topic modelling identified five topics. The largest (one-third of all tweets) was related to sports events. The remaining two-thirds could be allocated to four topics connected to communication about climate-related heat or heat waves. Two of these were on the impacts of heat and heat waves (health impacts 20%; social impacts 16%), one was on extreme weather and climate change attribution (17%) and the last one was on perceptions and warning (13%). The number of tweets in each week corresponded well with major heat wave occurrences in Argentina, Australia, the USA and South Asia (India and Pakistan), indicating that people posting tweets were aware of the threat from heat and its impacts on the society. Among the words frequently used within the topic ‘Social impacts’ were ‘air-conditioning’ and ‘electricity’, suggesting links between coping strategies and financial pressure. Apart from analysing the content of tweets, new insights were also obtained from analysing how people engaged with Twitter tweets about heat or heat waves. We found that tweets posted early, and which were then shared by other influential Twitter users, were among the most popular. Finally, we found that the most popular tweets belonged to individual scientists or respected news outlets, with no evidence that misinformation about climate change-related heat is widespread.

A Multimodal Ensemble Machine Learning Approach to COVID-19 Misinformation Detection in Twitter

The emergence of social media platforms has unquestionably altered the manner in which people ingest information, with tweets now functioning as the primary source for news and other types of content. However, the proliferation of false news on these platforms has become a major concern, as it poses a severe threat to both individuals and society as a whole. Consequently, it is crucial to develop efficient methods for detecting false news in tweets. This study presents a novel hybrid approach that integrates the textual content of tweets with auxiliary features to detect false news. Our approach uses a pre-trained transformer-based language model, COVID-twitter-BERT to encode the text content of tweets into a dense representation that captures their meaning. The auxiliary features, such as sentiment score, credibility score, engagement score, average retweet count, average favourite count, and average followers of followers, are fed into a stacking classifier-based model to predict the trustworthiness score of the tweet. By combining the predictions of both models, we demonstrate that our approach outperforms baseline methods, emphasising the significance of utilising both text content and auxiliary features for Twitter false news detection. Our research considerably advances the field of detecting false news by demonstrating the effectiveness of integrating transformer-based language models and machine learning models for this task. Our findings provide valuable insights for improving the detection of false news on social media.

1000 Days of COVID-19: A Gender-Based Long-Term Investigation into Attitudes With Regards to Vaccination

The coronavirus pandemic has undoubtedly been one of the major recent events that have affected our society at the global level. During this period, unprecedented measures have been imposed worldwide by authorities in an effort to contain the spread of the disease. These measures have led to a worldwide debate among the public, occurring not least within the forum of social media, tapping into pre-existing trends of skepticism, such as vaccine hesitancy. At the same time, it has become apparent that the pandemic affected women and men differently. With these two themes in view, the paper aims to analyze using a data-driven approach the evolution of opinions with regards to vaccination against COVID-19 throughout the entire duration of the pandemic from the point of view of gender. For this analysis, approximately 1,500,000 short user-contributed texts have been retrieved from the popular microblogging platform Twitter, posted between 30 January 2020 and 30 November 2022. Using a machine learning approach, several classifiers have been trained to identify the likely gender ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">female</i> or <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">male</i> ) of the author, as well as the stance of the specific post towards the COVID-19 vaccines ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">neutral</i> , <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">in favor</i> , or <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">against</i> ), achieving 85.69% and 93.64% weighted accuracy measures for each problem, respectively. Based on this analysis, it can be observed that most tweets exhibit a <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">neutral</i> stance, while the number of tweets <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">in favor</i> of vaccination is greater than the number of tweets opposing vaccination, with the distribution varying across time in response to specific events. The subject matter of the tweets varied more between stances than between genders, suggesting that there is no significant difference between the contents of tweets posted by females and males. We also find that while the overall engagement on Twitter with the topic of vaccination against COVID-19 is on the wane, there has been a rise in the number of <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">against</i> tweets continuing into the present.

Twitter trends in #Parasitology determined by text mining and topic modelling

This study investigated the emergence and use of Twitter, as of July 2023 being rebranded as X, as the main forum for social media communication in parasitology. A dataset of tweets was constructed using a keyword search of Twitter with the search terms ‘malaria’, ‘Plasmodium’, ‘Leishmania’, ‘Trypanosoma’, ‘Toxoplasma’ and ‘Schistosoma’ for the period from 2011 to 2020. Exploratory data analyses of tweet content were conducted, including language, usernames and hashtags. To identify parasitology topics of discussion, keywords and phrases were extracted using KeyBert and biterm topic modelling. The sentiment of tweets was analysed using VADER. The results show that the number of tweets including the keywords increased from 2011 (for malaria) and 2013 (for the others) to 2020, with the highest number of tweets being recorded in 2020. The maximum number of yearly tweets for Plasmodium, Leishmania, Toxoplasma, Trypanosoma and Schistosoma was recorded in 2020 (2804, 2161, 1570, 680 and 360 tweets, respectively). English was the most commonly used language for tweeting, although the percentage varied across the searches. In tweets mentioning Leishmania, only ∼37% were in English, with Spanish being more common. Across all the searches, Portuguese was another common language found. Popular tweets on Toxoplasma contained keywords relating to mental health including depression, anxiety and schizophrenia. The Trypanosoma tweets referenced drugs (benznidazole, nifurtimox) and vectors (bugs, triatomines, tsetse), while the Schistosoma tweets referenced areas of biology including pathology, eggs and snails. A wide variety of individuals and organisations were shown to be associated with Twitter activity. Many journals in the parasitology arena regularly tweet about publications from their journal, and professional societies promote activity and events that are important to them. These represent examples of trusted sources of information, often by experts in their fields. Social media activity of influencers, however, who have large numbers of followers, might have little or no training in science. The existence of such tweeters does raise cause for concern to parasitology, as one may start to question the quality of information being disseminated.

Multimodal Arabic Rumors Detection

Recently, the use of social media platforms has increased with ease of use and fast accessibility, making such platforms a place of rumor proliferation owing to the lack of posting constraints and content authentication. Therefore, there is a need to leverage artificial intelligence techniques to detect rumors on social media platforms to prevent their adverse effects on society and individuals. Most existing works that detect rumors in Arabic target the textual features of the tweet content. Nevertheless, tweets contain different types of content, such as (text, images, videos, and URLs), and the visual features of tweets play an essential role in rumor diffusion. This study proposes an Arabic rumor detection model to detect rumors on Twitter using textual and visual image features through two types of multimodal fusion: early and late fusion. In addition, we leveraged the transfer learning of the pre-trained language and vision models. Different experiments were conducted to select the best textual and visual feature extractors for building a multimodal model. MARBERTv2 was used as a textual feature extractor, whereas the ensemble of VGG-19 and ResNet50 was used as a visual feature extractor to build the multimodal model. Subsequently, the language and vision models of the single models were used as a baseline to compare their results with those of multimodal models. Finally, the experimental results demonstrate the effectiveness of textual features in rumor detection tasks compared to multimodal models.

Toward reduction of detrimental effects of hurricanes using a social media data analytic Approach: How climate change is perceived?

During natural disasters, there is a noticeably increased use of social media sites such as Twitter. Substantial research on social media data use during disasters has been conducted in the past decade since various social media platforms have emerged and gained popularity. This research highlights a thorough examination of the textual content of users’ posts shared on Twitter across the 48 contiguous U.S. states (CONUS) during hurricanes Harvey (2017) and Dorian (2019). We processed and analyzed 35 million tweets by classifying them into the main topics of concern discussed on Twitter over the CONUS. Sentiment analysis, topic modeling, and topic classification are a few of the Artificial Intelligence techniques from Natural Language Processing (NLP) that we employed in this work to analyze the Twitter data. Applying the NLP techniques on this large volume of data, made it possible to classify the tweet content into distinct categories in order to reveal valuable information on social response to hurricanes and assist crisis management agencies and disaster responders during and post disasters. Furthermore, this study offers helpful insights on the way climate change is discussed on Twitter before, during and after hurricane Harvey and Dorian. The outcome of this study uncovers detailed information on social response to hurricanes which benefits disaster managers and responders in reducing the detrimental effects of such extreme events and enhancing community readiness when these events occur.

Dünya Sağlık Örgütü’nün Covid-19 Pandemisi Öncesi ve Sırasında Twitter Kullanımı: Tweetlerin Duygu ve Metin Analizi

Social media as a form of mass self-communication offers a great deal of potential and opportunities for users and organizations to get, diffuse, redistribute, and monitor information and messages due to its features of accessibility, digitality, and time-space independence. Especially during a crisis, being able to harness the power of social media is an im- perative management practice not only to provide the public with accurate information but also to deter misleading claims and rumors regarding the subject. Based on this assumption, this study aims to measure and compare the overall sentiment of the World Health Organization’s (WHO) tweets before and during the coronavirus pandemic. Along with the overall senti- ment of tweets, the contents of tweets, and the WHO’s Twitter use practices before and during the pandemic were assessed in the framework of crisis management and health communication. Within the scope of our study, the WHO’s 34.673 tweets between 2018 and 2021 were examined with sentiment and textual analysis. The study also indicates the health policies be- fore and during the covid19 pandemic and crisis management tactics by analyzing and close reading the WHO’s tweets. The findings reveal that the WHO posted more tweets to inform the public and minimize infodemic while it deliberately preferred positive messages (words and sentiments) in its tweets during the pandemic.

Tweeting television between innovation and normalization: How Lebanese television and audiences are making use of Twitter in political talk shows

Traditional media have progressively integrated newer media practices, with the constant emergence of new digital technologies, without abandoning their former ones. The adoption of Twitter by TV channels and by other social actors during political talk shows is a case in point. This article aims to assess whether the hybridization of TV talk shows and Twitter has innovated or normalized existing patterns of communication. In the former case, by enabling more interaction between different actors and space for audience deliberation, or in the latter case, by reproducing a traditional one-way communication and a centralized network of information that remains controlled and oriented by the elite (journalists and politicians). The incorporation of older and emergent media logic has an impact on the construction and distribution of political information as well as on the power relationships between journalists, politicians, and TV audiences. Besides allowing political talk shows to expand their flow of information, and to promote their news online, hybridized practices have not only altered the way citizens consume and engage with political information but also how they counter-frame traditional political media content by producing new ones. The research methodology consists of descriptive, content, and network analyses of tweets collected from three Lebanese local TV political talk show “Sarelwa2et” (MTV), “Btefro’ aa Watan” (Al Jadeed), and “Vision 2030” (LBCI) between February and March 2022. Results revealed that TV talk shows are making use of Twitter as a top-down transmitter of information and resorting poorly to its interactive potential. Some newer media practices of Twitter, such as @mention and replies are being applied but only to interact with politicians and journalists, failing to engage with a larger array of voices and thus, leading to an elite-centric discourse within the network. Also, tweets are mostly used to inform audiences and promote TV programs. In addition, network analyses of talk shows via hashtags demonstrated the central and not monopolized role of politicians and journalists as influencers, bridges, and quick spreaders of information. Finally, content analysis of dual screeners’ tweets (n=6000) indicated very little space in a Habermasian public sphere. The total of subjective opinions, irony, and attack/insult tweets are still higher than the total of the introduction of new issues and counter-frames tweets.

Analysis of user's car parking behaviour through twitter hashtags

AbstractDue to its ease and popularity, social media has recently become an essential source of data for researchers and various stakeholder groups that seek a reliable assessment of their policies based on a comprehensive understanding of users' feedback and inputs, which are reflected in their posts and discussions. In this study, we investigate the issue of users' car parking behaviour through a comprehensive analysis of related hashtags collected from the Twitter social media platform. For this purpose, we adopted a two‐step strategy where in the first stage, a surface‐level analysis of the identified hashtags involving inductive reasoning, sentiment analysis, and user interaction in terms of engagement and diversity scores is performed. In the second phase, a tweet content analysis is performed using sentiment analysis and Empath categorization with respect to the most frequent wordings (assimilating to separate topics), in the same spirit as aspect‐sentiment analysis, to gain further insights regarding the occurrence of negative and positive posts. A quantitative evaluation of the coherence of the Empath categorization indicates that achieved 0.83% coherence score and outperformed both LDA and LSA that had a score of 0.77% and 0.76% respectively. Furthermore, the common word technique assimilated to aspect sentiment compared to the state‐of‐the‐art model for aspect sentiment‐based deberta‐v3‐base model. Besides, the influence of bots or spammers is evaluated using engagement/diversity measures and Botometer API. The results provide valuable insights in terms of discriminating between positive and negative posts and the correlation of surface‐level analysis with content‐based analysis, as well as the impact of various categorizations. The results expect to enable urban planners and policymakers to advance evidence‐based policing in the future design of intelligent parking systems.

Temporal intensity modeling of sub‐events in microblogs

SummarySocial media has evolved into a conventional source of obtaining and tracking news events in the digital world. Most news events constitute a flow of sub‐events, that are distinct and independent events, related to the same base news event. These sub‐events have the potential to alter the course of the news event and also the emotional responses toward the news event entirely. Discovering the intensity of user attention toward sub‐events at set points of time, aids in modeling the evolution of sub‐events and consecutively, the news event as a whole. This work proposes two methodologies for day‐to‐day temporal analysis of sub‐events: a top‐down approach based on the significance of hashtags and a bottom‐up approach based on the combined intensity of components such as hashtags and tweet content. In comparison, the approach based on the combined intensity of components proves very promising in representing the temporal intensity of sub‐events.