Content Of Tweets Research Articles

Look at this! Radical right-wing populist information sharing practices on Twitter in Portugal, Spain, and Brazil

Politicians communicate on social media not only by direct utterances, but also by where they choose to direct followers’ attention using strategic interaction, audiovisual content and outlinks, i.e. their information sharing practices. Political communication research has focused on the former, but the latter is a relevant way to foster advantageous informational environments and interact with media systems. This is especially important for radical right-wing populists, due to their reliance on social media and adjacency to disinformation. Portugal, Spain, and Brazil provide important research cases due to media system characteristics and the different positions within the political system of the relevant leaders, which generates different opportunity structures. This article approaches relevant radical right-wing populist leaders’ information sharing practices by analysing the non-textual content of tweets ( n = 6087), emphasizing how they reflect relations with local media systems.

Stroke recovery—what are people talking about on Twitter? A content analysis

Purpose Prioritisation exercises seek out what matters to key stakeholders to inform the planning of research. Social media platforms are potentially useful data sources. The aim was to examine the content of tweets, short messages containing text and pictures, to ascertain the priorities of Twitter users regarding stroke recovery. Materials and methods Content analysis of Twitter was conducted. An electronic search used the identifiers: #strokesurvivor and #strokerecovery. Tweets spanning four weeks from January 2021 were analysed. Results There were 1361 tweets extracted and 486 analysed following exclusion of duplicates and unrelated material. Six themes were uncovered (n = number of tweets): maintaining motivation and positivity (153); sharing of resources (146); raising awareness of stroke (74); symptomatic aspects of recovery (39); experience of rehabilitation (63); and concerns about Covid-19 (17). Conclusions Despite the brevity of tweets, a rich picture arose. A key limitation was lack of biographical data about Twitter users. Recommendations about topics requiring attention from stroke researchers, clinicians and policy makers are: management of psychological problems; public perception of stroke; rehabilitation considerations including treatment burden, person-centred care and equality of care; symptom management including fatigue and aphasia. Findings can be used to supplement and validate other priority-setting exercises.

Herding and investor sentiment after the cryptocurrency crash: evidence from Twitter and natural language processing

Although the 2022 cryptocurrency market crash prompted despair among investors, the rallying cry, “wagmi” (We’re all gonna make it.) emerged among cryptocurrency enthusiasts in the aftermath. Did cryptocurrency enthusiasts respond to this crash differently compared to traditional investors? Using natural language processing techniques applied to Twitter data, this study employed a difference-in-differences method to determine whether the cryptocurrency market crash had a differential effect on investor sentiment toward cryptocurrency enthusiasts relative to more traditional investors. The results indicate that the crash affected investor sentiment among cryptocurrency enthusiastic investors differently from traditional investors. In particular, cryptocurrency enthusiasts’ tweets became more neutral and, surprisingly, less negative. This result appears to be primarily driven by a deliberate, collectivist effort to promote positivity within the cryptocurrency community (“wagmi”). Considering the more nuanced emotional content of tweets, it appears that cryptocurrency enthusiasts expressed less joy and surprise in the aftermath of the cryptocurrency crash than traditional investors. Moreover, cryptocurrency enthusiasts tweeted more frequently after the cryptocurrency crash, with a relative increase in tweet frequency of approximately one tweet per day. An analysis of the specific textual content of tweets provides evidence of herding behavior among cryptocurrency enthusiasts.

EXPRESS: Corporate Use of Social Media after ESG Incidents

Despite increasing investments in Environmental, Social, and Governance (ESG) initiatives and practices, firms often fail to meet public expectations, causing ESG incidents. While firms often choose to remain silent after an incident, we argue that this is not attractive to firms anymore on social media platforms, where consumers and stakeholders can freely share information and concerns. That is, firms tend to use official social media accounts to increase communication frequency and communicate with stakeholders about the incidents (i.e., incident-related posts) after the occurrences. We are also interested in the extent to which firms would adjust their use of social media in terms of non-incident-related posts, as the current prevailing practical advice and studies offer contradicting predictions. Using data from different sources, we construct an event-based firm-day dataset and empirically show that firms significantly increase the number of social media posts after ESG incidents. The impact is more salient for firms in consumer-oriented industries and when the incident is more impactful. Using a semi-supervised, dictionary-based approach, we delve into the content of tweets and demonstrate that firms are inclined to increase both the number of incident-related and the number of non-incident-related tweets after an ESG incident. The follow-up analyses at the incident level indicate that firms that post more after an ESG incident experience a better reaction from the capital market, especially for customer-oriented firms or incidents that receive high attention from the traditional media.

Twitter Discussions on #digitaldementia: Content and Sentiment Analysis.

Digital dementia is a term that describes a possible decline in cognitive abilities, especially memory, attributed to the excessive use of digital technology such as smartphones, computers, and tablets. This concept has gained popularity in public discourse and media lately. With the increasing use of social media platforms such as Twitter (subsequently rebranded as X), discussions about digital dementia have become more widespread, which offer a rich source of information to understand public perceptions, concerns, and sentiments regarding this phenomenon. The aim of this research was to delve into a comprehensive content and sentiment analysis of Twitter discussions regarding digital dementia using the hashtag #digitaldementia. Retrospectively, publicly available English-language tweets with hashtag combinations related to the topic of digital dementia were extracted from Twitter. The tweets were collected over a period of 15 years, from January 1, 2008, to December 31, 2022. Content analysis was used to identify major themes within the tweets, and sentiment analysis was conducted to understand the positive and negative emotions associated with these themes in order to gain a better understanding of the issues surrounding digital dementia. A one-way ANOVA was performed to gather detailed statistical insights regarding the selected tweets from influencers within each theme. This study was conducted on 26,290 tweets over 15 years by 5123 Twitter users, mostly female users in the United States. The influencers had followers ranging from 20,000 to 1,195,000 and an average of 214,878 subscribers. The study identified four themes regarding digital dementia after analyzing tweet content: (1) cognitive decline, (2) digital dependency, (3) technology overload, and (4) coping strategies. Categorized according to Glaser and Strauss's classifications, most tweets (14,492/26,290, 55.12%) fell under the categories of wretched (purely negative) or bad (mostly negative). However, only a small proportion of tweets (3122/26,290, 11.86%) were classified as great (purely positive) or swell sentiment (mostly positive). The ANOVA results showed significant differences in mean sentiment scores among the themes (F3,3581=29.03; P<.001). The mean sentiment score was -0.1072 (SD 0.4276). Various negative tweets have raised concerns about the link between excessive use of digital devices and cognitive decline, often known as digital dementia. Of particular concern is the rapid increase in digital device use. However, some positive tweets have suggested coping strategies. Engaging in digital detox activities, such as increasing physical exercise and participating in yoga and meditation, could potentially help prevent cognitive decline.

Moral panics on social media are fueled by signals of virality.

Moral panics have regularly erupted in society, but they appear almost daily on social media. We propose that social media helps fuel moral panics by combining perceived societal threats with a powerful signal of social amplification-virality. Eight studies with multiple methods test a social amplification model of moral panics in which virality amplifies perceptions of threats posed by deviant behavior and ideas, prompting moral outrage expression. Three naturalistic studies of Twitter (N = 237,230) reveal that virality predicts moral outrage in response to tweets about controversial issues, even when controlling for specific tweet content. Five experiments (N = 1,499) reveal the causal impact of virality on outrage expression and suggest that feelings of danger mediate this effect. This work connects classic ideas about moral panics with ongoing research on social media and provides a perspective on the nature of moral outrage. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

The Sharing of Pandemic-Related Information From U.S. Government Twitter Accounts

The rapid onset of COVID-19 created a desire for information about the virus and government-directed response efforts. This study explores the sharing of Twitter posts from U.S. government officials and organizations during the onset of the COVID-19 pandemic. This study followed two paths of investigation. First, significant differences in the retweet frequency of government sources were determined based on the source (public officials and agencies, Republican- or Democrat-led states) and differentiated by equal time periods (40 days pre- or post-pandemic declaration). The study findings show that the frequency at which government accounts were retweeted in the U.S. was significantly impacted by the pandemic period, with the post-pandemic period significantly more active as governments wrestled with unprecedented economic, policy, and societal concerns related to the crisis. A greater reliance on public officials as opposed to public agencies or departments was seen, as was a greater tendency to retweet the posts of government officials from Democrat-led states. The second path examined the content of government tweets to determine the linguistic factors which predict retweet frequency. Study results found that both positive and negative emotional language in tweets predicted the frequency of retweeting government sources, though the results varied based on the source. The study results provide insight into the government-affiliated information sources users chose to share with their networks as well as the types of language that appeal to citizens during times of crisis.

How Do Teachers Collaborate in Informal Professional Learning Activities? An Epistemic Network Analysis

Teachers turn to many sources for support and professional learning, including social media-based communities that have shown promise to help teachers access resources and facilitate productive exchanges. Although such online communities show promise, questions about their quality for providing a suitable learning environment remain insufficiently answered. In this study, we examine how teachers’ engagement on Twitter (now known as “X”) may adhere to characteristics of high-quality professional development (PD) activities. In that, we employ advanced conversational analysis techniques that extend the primarily descriptive methods used in prior research. Specifically, we collected data from three Twitter communities related to Advanced Placement Biology (N = 2,040 tweets, N = 93 teachers). Qualitative two-cycle content analyses derived both tweet content and sentiment. Using epistemic network analyses, we examined the collaborative structures to examine how participation patterns can identify characteristics of high-quality online PD. Results indicate that some teachers use Twitter with a content focus and coherent to their individual contexts and prior knowledge. Notably, differences in collaboration and participation patterns by teachers’ overall activity level hint at the existence of an online community of practice. More active teachers communicated more about how their individual contexts relate to instruction, whereas less active teachers exhibited more targeted engagement, for instance, related to sharing teaching resources and organizing learning opportunities. Overall, this study illustrates how Twitter may provide a meaningful learning environment to teachers so that it can serve as a relevant avenue for their professional learning.

Nonrandom Tweet Mortality and Data Access Restrictions: Compromising the Replication of Sensitive Twitter Studies

Abstract Used by politicians, journalists, and citizens, Twitter has been the most important social media platform to investigate political phenomena such as hate speech, polarization, or terrorism for over a decade. A high proportion of Twitter studies of emotionally charged or controversial content limit their ability to replicate findings due to incomplete Twitter-related replication data and the inability to recrawl their datasets entirely. This paper shows that these Twitter studies and their findings are considerably affected by nonrandom tweet mortality and data access restrictions imposed by the platform. While sensitive datasets suffer a notably higher removal rate than nonsensitive datasets, attempting to replicate key findings of Kim’s (2023, Political Science Research and Methods 11, 673–695) influential study on the content of violent tweets leads to significantly different results. The results highlight that access to complete replication data is particularly important in light of dynamically changing social media research conditions. Thus, the study raises concerns and potential solutions about the broader implications of nonrandom tweet mortality for future social media research on Twitter and similar platforms.

Pediatric Cancer Communication on Twitter: Natural Language Processing and Qualitative Content Analysis.

During the COVID-19 pandemic, Twitter (recently rebranded as "X") was the most widely used social media platform with over 2 million cancer-related tweets. The increasing use of social media among patients and family members, providers, and organizations has allowed for novel methods of studying cancer communication. This study aimed to examine pediatric cancer-related tweets to capture the experiences of patients and survivors of cancer, their caregivers, medical providers, and other stakeholders. We assessed the public sentiment and content of tweets related to pediatric cancer over a time period representative of the COVID-19 pandemic. All English-language tweets related to pediatric cancer posted from December 11, 2019, to May 7, 2022, globally, were obtained using the Twitter application programming interface. Sentiment analyses were computed based on Bing, AFINN, and NRC lexicons. We conducted a supplemental nonlexicon-based sentiment analysis with ChatGPT (version 3.0) to validate our findings with a random subset of 150 tweets. We conducted a qualitative content analysis to manually code the content of a random subset of 800 tweets. A total of 161,135 unique tweets related to pediatric cancer were identified. Sentiment analyses showed that there were more positive words than negative words. Via the Bing lexicon, the most common positive words were support, love, amazing, heaven, and happy, and the most common negative words were grief, risk, hard, abuse, and miss. Via the NRC lexicon, most tweets were categorized under sentiment types of positive, trust, and joy. Overall positive sentiment was consistent across lexicons and confirmed with supplemental ChatGPT (version 3.0) analysis. Percent agreement between raters for qualitative coding was 91%, and the top 10 codes were awareness, personal experiences, research, caregiver experiences, patient experiences, policy and the law, treatment, end of life, pharmaceuticals and drugs, and survivorship. Qualitative content analysis showed that Twitter users commonly used the social media platform to promote public awareness of pediatric cancer and to share personal experiences with pediatric cancer from the perspective of patients or survivors and their caregivers. Twitter was frequently used for health knowledge dissemination of research findings and federal policies that support treatment and affordable medical care. Twitter may serve as an effective means for researchers to examine pediatric cancer communication and public sentiment around the globe. Despite the public mental health crisis during the COVID-19 pandemic, overall sentiments of pediatric cancer-related tweets were positive. Content of pediatric cancer tweets focused on health and treatment information, social support, and raising awareness of pediatric cancer.

Social-media-based crisis communication: Assessing the engagement of local agencies in Twitter during Hurricane Irma

As social media platforms have become vital means of communication, it has become imperative for emergency managers and policy makers to understand how people are interacting with different agencies on these platforms for enhancing community response coordination during disasters. Although many public agencies have already adopted social media platforms for crisis communication purposes, empirical evidence on whether and how these agencies are effectively engaged on these platforms is lacking. This research aims to examine crisis communication activities of a variety of agencies in Twitter (now X) in response to Hurricane Irma in 2017. The study analyzed 13,353 hurricane-related tweets posted by the local agencies from eight counties in East Central Florida as well as federal and state agencies and other nonprofit and private organizations during the hurricane response period. An engagement metric was applied on these tweets to determine which agencies were most active in disseminating information during Hurricane Irma. The results revealed the most engaged local agencies in Twitter during Hurricane Irma for crisis communication and the Emergency Support Functions (ESFs) associated with them in coordinating disaster response activities. The study further investigated Twitter profiles and relevant attributes of these actively engaged agencies, and the contents of the hurricane-related tweets during Hurricane Irma. The study found that local agencies remain active and engaged in Twitter during a disaster compared to other levels of agencies. The ESF representing communication (ESF #2) was one of the most frequent ESFs associated with these active agencies; however, ESF #6 (Mass Care and Sheltering) was also found to be highly utilized. The results also provided insights on crisis communication performance of these agencies in terms of the three dimensions of social media engagement including popularity, commitment, and virality, which are correlated to the counts of like, reply, and retweet of each post, respectively. The study includes recommendations to local government and partner organizations and emergency managers to improve crisis communication in social media and suggests future research directions.

Towards understanding the role of content-based and contextualized features in detecting abuse on Twitter

This paper presents a novel approach for detecting abuse on Twitter. Abusive posts have become a major problem for social media platforms like Twitter. It is important to identify abuse to mitigate its potential harm. Many researchers have proposed methods to detect abuse on Twitter. However, most of the existing approaches for detecting abuse look only at the content of the abusive tweet in isolation and do not consider its contextual information, particularly the tweets posted before the abusive tweet. In this paper, we propose a new method for detecting abuse that uses contextual information from the tweets that precede and follow the abusive tweet. We hypothesize that this contextual information can be used to better understand the intent of the abusive tweet and to identify abuse that content-based methods would otherwise miss. We performed extensive experiments to identify the best combination of features and machine learning algorithms to detect abuse on Twitter. We test eight different machine learning classifiers on content- and context-based features for the experiments. The proposed method is compared with existing abuse detection methods and achieves an absolute improvement of around 7%. The best results are obtained by combining the content and context-based features. The highest accuracy of the proposed method is 86%, whereas the existing methods used for comparison have highest accuracy of 79.2%.

Sentiment analysis of tweet content on Hurricane Dorian: Sensemaking in digital journalistic inquiry ecology

Twitter is a powerful digital journalistic instrument and evidence shows journalists were transferring authority to Twitter. With journalistic information ecology becoming imbalanced, it is valuable to research how journalists may use Twitter to discover accurate and reliable information and maintain a vast overview of news events without shifting the power as the fourth estate. The purpose of this study is to provide a possible digital journalistic inquiry model to identify trending topics, distinguish reliable journalistic information while maintaining the balance of journalistic information ecology. Utilizing a large-scale dataset – 1.2 million tweets collected from Twitter API – this study executed cutting-edge network analysis and sentiment analysis to fill in the knowledge gap through a case study on Hurricane Dorian. The study found that the impact of traditional opinion leaders on information diffusion is declining. On the contrary, top in-degree centrality users play more important roles in information diffusion on Twitter. Moreover, tweets with negative polarity opinions were retweeted more. In addition, non-opinion leaders’ negatively polarized tweets were retweeted more than positively polarized ones, although it is not the same case with opinion leaders. With the change of journalistic ecology, identifying top in-degree centrality users and examining their tweets will provide useful resources for journalists to identify keywords, trending themes and predict how likely a topic may interest audience based on degree of polarity and number of retweets on Twitter. The results provide useful patterns for journalists to follow in sensemaking tasks in digital journalistic inquiry.

Using Natural Language Processing to Explore Social Media Opinions on Food Security: Sentiment Analysis and Topic Modeling Study.

Social media has the potential to be of great value in understanding patterns in public health using large-scale analysis approaches (eg, data science and natural language processing [NLP]), 2 of which have been used in public health: sentiment analysis and topic modeling; however, their use in the area of food security and public health nutrition is limited. This study aims to explore the potential use of NLP tools to gather insights from real-world social media data on the public health issue of food security. A search strategy for obtaining tweets was developed using food security terms. Tweets were collected using the Twitter application programming interface from January 1, 2019, to December 31, 2021, filtered for Australia-based users only. Sentiment analysis of the tweets was performed using the Valence Aware Dictionary and Sentiment Reasoner. Topic modeling exploring the content of tweets was conducted using latent Dirichlet allocation with BigML (BigML, Inc). Sentiment, topic, and engagement (the sum of likes, retweets, quotations, and replies) were compared across years. In total, 38,070 tweets were collected from 14,880 Twitter users. Overall, the sentiment when discussing food security was positive, although this varied across the 3 years. Positive sentiment remained higher during the COVID-19 lockdown periods in Australia. The topic model contained 10 topics (in order from highest to lowest probability in the data set): "Global production," "Food insecurity and health," "Use of food banks," "Giving to food banks," "Family poverty," "Food relief provision," "Global food insecurity," "Climate change," "Australian food insecurity," and "Human rights." The topic "Giving to food banks," which focused on support and donation, had the highest proportion of positive sentiment, and "Global food insecurity," which covered food insecurity prevalence worldwide, had the highest proportion of negative sentiment. When compared with news, there were some events, such as COVID-19 support payment introduction and bushfires across Australia, that were associated with high periods of positive or negative sentiment. Topics related to food insecurity prevalence, poverty, and food relief in Australia were not consistently more prominent during the COVID-19 pandemic than before the pandemic. Negative tweets received substantially higher engagement across 2019 and 2020. There was no clear relationship between topics that were more likely to be positive or negative and have higher or lower engagement, indicating that the identified topics are discrete issues. In this study, we demonstrated the potential use of sentiment analysis and topic modeling to explore evolution in conversations on food security using social media data. Future use of NLP in food security requires the context of and interpretation by public health experts and the use of broader data sets, with the potential to track dimensions or events related to food security to inform evidence-based decision-making in this area.

A novel fine-grained rumor detection algorithm with attention mechanism

Rumors circulating on social media platforms have consistently represented a substantial threat to societal security and stability. Both academia and the industry have dedicated heightened focus to addressing the issue of rumor detection. Recent research has made significant progress in using deep neural networks to model the textual content and propagation structure of rumors. However, these methods model rumor-related features at a coarse-grained level and do not take full advantage of the various contextual information associated with rumors. In this paper, we propose a new model called Hybrid Rumor Detection Model with Co-Attention Mechanisms(CoAHRD), which utilizes the original tweet content, social context information, and user information for rumor detection. First, we use the Fine-Grained Feature Learning(FGFL) algorithm to extract fine-grained features from the tweets. Based on this, we use the Graph Convolutional Network(GCN) to learn the Multi-relation graph propagation structure features of rumors. Next, we combine FGFL features with temporal encoding information to model the temporal structure of rumors. Then, we introduce a User Feature-Based Co-Attention Network(UFCoAN) to learn the tendency of different users to spread rumors. Finally, we fuse the above features through a fully connected layer and perform rumor detection. Extensive experiments on two publicly available datasets, PHEME and TWITTER15, show that our method outperforms the current mainstream methods. In particular, in terms of accuracy, our model improves by 0.9% and 1.6% over the best baseline method on the Twitter15 and PHEME datasets, respectively.

A two-stage framework for Arabic social media text misinformation detection combining data augmentation and AraBERT

Misinformation can profoundly impact the reputation of an entity, and eliminating its spread has become a critical concern across various applications. Social media, often a primary source of information, can significantly influence individuals’ perspectives through content from less credible sources. The utilization of machine-learning (ML) algorithms can facilitate automated, large-scale analysis of textual content, contributing to the rapid and efficient processing of extensive datasets for informed decision-making. Since the performance of ML models is highly affected by the size of the training data, many research papers have presented different approaches to solve the problem of limited dataset size. The data augmentation (DA) approach is one of these strategies, aiming to enhance ML model performance by increasing the amount of training data. DA generates new instances by applying different transformations to the original data instances. While many DA techniques have been investigated for various languages, such as English, achieving an enhancement of the classification model’s performance on the new augmented dataset compared to the original dataset, there is a lack of studies on the Arabic language due to its unique characteristics. This paper introduces a novel two-stage framework designed for the automated identification of misinformation in Arabic textual content. The first stage aims to identify the optimal representation of features before feeding them to the ML model. Diverse representations of tweet content are explored, including N-grams, content-based features, and source-based features. The second stage focuses on investigating the DA effect through the back-translation technique applied to the original training data. Back-translation entails translating sentences from the target language (in this case, Arabic) into another language and then back to Arabic. As a result of this procedure, new examples for training are created by introducing variances in the text. The study utilizes support vector machine (SVM), naive Bayes, logistic regression (LR), and random forest (RF) as baseline algorithms. Additionally, AraBERT transformer pre-trained language models are used to relate the instance’s label and feature representation of the input. Experimental outcomes demonstrate that misinformation detection, coupled with data augmentation, enhances accuracy by a noteworthy margin 5 to 12% compared to baseline machine-learning algorithms and pre-trained models. Remarkably, the results show the superiority of the N-grams approach over traditional state-of-the-art feature representations concerning accuracy, recall, precision, and F-measure metrics. This suggests a promising avenue for improving the efficacy of misinformation detection mechanisms in the realm of Arabic text analysis.

Use of Large Language Models to Assess the Likelihood of Epidemics From the Content of Tweets: Infodemiology Study.

Previous work suggests that Google searches could be useful in identifying conjunctivitis epidemics. Content-based assessment of social media content may provide additional value in serving as early indicators of conjunctivitis and other systemic infectious diseases. We investigated whether large language models, specifically GPT-3.5 and GPT-4 (OpenAI), can provide probabilistic assessments of whether social media posts about conjunctivitis could indicate a regional outbreak. A total of 12,194 conjunctivitis-related tweets were obtained using a targeted Boolean search in multiple languages from India, Guam (United States), Martinique (France), the Philippines, American Samoa (United States), Fiji, Costa Rica, Haiti, and the Bahamas, covering the time frame from January 1, 2012, to March 13, 2023. By providing these tweets via prompts to GPT-3.5 and GPT-4, we obtained probabilistic assessments that were validated by 2 human raters. We then calculated Pearson correlations of these time series with tweet volume and the occurrence of known outbreaks in these 9 locations, with time series bootstrap used to compute CIs. Probabilistic assessments derived from GPT-3.5 showed correlations of 0.60 (95% CI 0.47-0.70) and 0.53 (95% CI 0.40-0.65) with the 2 human raters, with higher results for GPT-4. The weekly averages of GPT-3.5 probabilities showed substantial correlations with weekly tweet volume for 44% (4/9) of the countries, with correlations ranging from 0.10 (95% CI 0.0-0.29) to 0.53 (95% CI 0.39-0.89), with larger correlations for GPT-4. More modest correlations were found for correlation with known epidemics, with substantial correlation only in American Samoa (0.40, 95% CI 0.16-0.81). These findings suggest that GPT prompting can efficiently assess the content of social media posts and indicate possible disease outbreaks to a degree of accuracy comparable to that of humans. Furthermore, we found that automated content analysis of tweets is related to tweet volume for conjunctivitis-related posts in some locations and to the occurrence of actual epidemics. Future work may improve the sensitivity and specificity of these methods for disease outbreak detection.

Sentence embedding to improve rumour detection performance model

&lt;span lang="EN-US"&gt;Recently, most individuals have preferred accessing the most recent news via social media platforms like Twitter as their primary source of information. Moreover, Twitter enables users to post and distribute tweets quickly and unsupervised. As a result, Twitter has become a popular platform for disseminating false information, such as rumours. These rumours were then propagated as accurate and influenced public opinion and decision-making. The issue will arise when a decision or policy with substantial consequences is made based on rumours. To avoid the negative impacts of rumours, several researchers have attempted to detect them automatically as early as feasible. Previous studies employed supervised learning methods to identify Twitter rumours and relied on feature extraction algorithms to extract tweet content and context elements. However, manually extracting features is time-consuming and labour-intensive. To encode each tweet's sentence as a vector based on its contextual meaning, we proposed utilising Bidirectional Encoder Representation of Transformer (BERT) as a sentence embedding. We then used these vectors to train some classifier models to detect rumours. Finally, we compared the performance of BERT-based models to feature engineering-based models. We discovered that the suggested BERT-based model improved all parameters by around 10% compared to the feature engineering-based classification model.&lt;/span&gt;

BERT for Twitter Sentiment Analysis: Achieving High Accuracy and Balanced Performance

The Bidirectional Encoder Representations from Transformers (BERT) model is used in this work to analyse sentiment on Twitter data. A Kaggle dataset of manually annotated and anonymized COVID-19-related tweets was used to refine the model. Location, tweet date, original tweet content, and sentiment labels are all included in the dataset. When compared to the Multinomial Naive Bayes (MNB) baseline, BERT's performance was assessed, and it achieved an overall accuracy of 87% on the test set. The results indicated that for negative feelings, the accuracy was 0.93, the recall was 0.84, and the F1-score was 0.88; for neutral sentiments, the precision was 0.86, the recall was 0.78, and the F1-score was 0.82; and for positive sentiments, the precision was 0.82, the recall was 0.94, and the F1-score was 0.88. The model's proficiency with the linguistic nuances of Twitter, including slang and sarcasm, was demonstrated. This study also identifies the flaws of BERT and makes recommendations for future research paths, such as the integration of external knowledge and alternative designs.

Cross sectional study of Twitter (X) use among academic anesthesiology departments in the United States.

Twitter (recently renamed X) is used by academic anesthesiology departments as a social media platform for various purposes. We hypothesized that Twitter (X) use would be prevalent among academic anesthesiology departments and that the number of tweets would vary by region, physician faculty size, and National Institutes of Health (NIH) research funding rank. We performed a descriptive study of Twitter (X) use by academic anesthesiology departments (i.e. those with a residency program) in 2022. Original tweets were collected using a Twitter (X) analytics tool. Summary statistics were reported for tweet number and content. The median number of tweets was compared after stratifying by region, physician faculty size, and NIH funding rank. Among 166 academic anesthesiology departments, there were 73 (44.0%) that had a Twitter (X) account in 2022. There were 3,578 original tweets during the study period and the median number of tweets per department was 21 (25th-75th = 0, 75) with most tweets (55.8%) announcing general departmental news and a smaller number highlighting social events (12.5%), research (11.1%), recruiting (7.1%), DEI activities (5.2%), and trainee experiences (4.1%). There was no significant difference in the median number of tweets by region (P = 0.81). The median number of tweets differed significantly by physician faculty size (P<0.001) with larger departments tweeting more and also by NIH funding rank (P = 0.005) with highly funded departments tweeting more. In 2022, we found that less than half of academic anesthesiology departments had a Twitter (X) account, and the median number of annual tweets per account was relatively low. Overall, Twitter (X) use was less common than anticipated among academic anesthesiology departments and most tweets focused on promotion of departmental activities or individual faculty. There may be opportunities for more widespread and effective use of Twitter (X) by academic anesthesiology departments including education about anesthesiology as a specialty.