Twitter Content Research Articles

RuSentiTweet: a sentiment analysis dataset of general domain tweets in Russian.

The Russian language is still not as well-resourced as English, especially in the field of sentiment analysis of Twitter content. Though several sentiment analysis datasets of tweets in Russia exist, they all are either automatically annotated or manually annotated by one annotator. Thus, there is no inter-annotator agreement, or annotation may be focused on a specific domain. In this article, we present RuSentiTweet, a new sentiment analysis dataset of general domain tweets in Russian. RuSentiTweet is currently the largest in its class for Russian, with 13,392 tweets manually annotated with moderate inter-rater agreement into five classes: Positive, Neutral, Negative, Speech Act, and Skip. As a source of data, we used Twitter Stream Grab, a historical collection of tweets obtained from the general Twitter API stream, which provides a 1% sample of the public tweets. Additionally, we released a RuBERT-based sentiment classification model that achieved F1 = 0.6594 on the test subset.

CHALLENGES OF TOURISM CAMPAIGNS IN THE NEW NORMAL ERA: ANALYSIS ON INDONESIAN GOVERNMENT'S SOCIAL MEDIA

The COVID-19 pandemic has had a severe impact on the tourism industry sector. The government is trying to revive the Indonesian tourism sector by issuing a new normal policy.Therefore, this study aims to determine how Twitter is used as a tourism campaign tool by the Indonesian government in the new normal era of the COVID-19 pandemic. This study uses a descriptive qualitative approach to explain the tourism campaign in the new normal era of the COVID-10 pandemic by the Indonesian government through Twitter social media. This study uses the NVIVO 12 Plus analysis technique with chart features, word frequency, and cluster analysis. The source of this research data is the Twitter account of the Ministry of tourism and creative economy and the charm of Indonesia. The findings in this study are the Twitter content of the Ministry of tourism and creative economy and the charm of Indonesia in campaigning for Indonesian tourism is about tourist facilities, tourist events, tourist destinations, and health protocol. The narrative of the Twitter account of the Ministry of Tourism and Creative Economy and Indonesian charm is related to invitations to travel to Indonesia, the beauty of tourism, and Indonesian tourist destinations. Actors involved in the Indonesian tourism campaign are Minister Sandiaga Salahudin Uno and Deputy Minister Angela Tanoesoedibjo, and President Joko Widodo. The two Twitter accounts have a balanced intensity in campaigning for Indonesian tourism from July 2020 to June 2021.

Comparing Emotional Valence Scores of Twitter Posts from Manual Coding and Machine Learning Algorithms to Gain Insights to Refine Interventions for Family Caregivers of Persons with Dementia.

We randomly extracted Korean-language Tweets mentioning dementia/Alzheimer's disease (n= 12,413) from November 28 to December 9, 2020. We independently applied three machine learning algorithms (Afinn, Syuzhet, and Bing) using natural language processing (NLP) techniques and qualitative manual scoring to assign emotional valence scores to Tweets. We then compared the means and distributions of the four emotional valence scores. Visual examination of the graphs produced indicated that each method exhibited unique patterns. The aggregated mean emotional valence scores from the NLP methods were mostly neutral, vs. slightly negative for manual coding (Afinn 0.029, 95% CI [-0.019, 0.077]; Syuzhet 0.266, [0.236, 0.295]; Bing -0.271, [-0.289, -0.252]; manual coding -1.601, [-1.632, -1.569]). One-way analysis of variance (ANOVA) showed no statistically significant differences among the four means after normalization. These findings suggest that the application of NLP can be fairly effective in extracting emotional valence scores from Korean-language Twitter content to gain insights regarding family caregiving for a person with dementia.

Data-driven strategies in operation management: mining user-generated content in Twitter

In recent years, the business ecosystem has focused on understanding new ways of automating, collecting, and analyzing data in order to improve products and business models. These actions allow operations management to improve prediction, value creation, optimization, and automatization. In this study, we develop a novel methodology based on data-mining techniques and apply it to identify insights regarding the characteristics of new business models in operations management. The data analyzed in the present study are user-generated content from Twitter. The results are validated using the methods based on Computer-Aided Text Analysis. Specifically, a sentimental analysis with TextBlob on which experiments are performed using vector classifier, multinomial naïve Bayes, logistic regression, and random forest classifier is used. Then, a Latent Dirichlet Allocation is applied to separate the sample into topics based on sentiments to calculate keyness and p-value. Finally, these results are analyzed with a textual analysis developed in Python. Based on the results, we identify 8 topics, of which 5 are positive (Automation, Data, Forecasting, Mobile accessibility and Employee experiences), 1 topic is negative (Intelligence Security), and 2 topics are neutral (Operational CRM, Digital teams). The paper concludes with a discussion of the main characteristics of the business models in the OM sector that use DDI. In addition, we formulate 26 research questions to be explored in future studies.

Differences in Suicide-Related Twitter Content According to User Influence.

Background: The content of suicide-specific social media posts may impact suicide rates, and putatively harmful and/or protective content may vary by the author's influence. Aims: This study sought to characterize how suicide-related Twitter content differs according to user influence. Method: Suicide-related tweets from July 1, 2015, to June 1, 2016, geolocated to Toronto, Canada, were collected and randomly selected for coding (n = 2,250) across low, medium, or high user influence levels (based on the number of followers, tweets, retweets, and posting frequency). Logistic regression was used to identify differences by user influence for various content variables. Results: Low- and medium-influence users typically tweeted about personal experiences with suicide and associations with mental health and shared morbid humor/flippant tweets. High-influence users tended to tweet about suicide clusters, suicide in youth, older adults, indigenous people, suicide attempts, and specific methods. Tweets across influence levels predominantly focused on suicide deaths, and few described suicidal ideation or included helpful content. Limitations: Social media data were from a single location and epoch. Conclusion: This study demonstrated more problematic content vis-à-vis safe suicide messaging in tweets by high-influence users and a paucity of protective content across all users. These results highlight the need for further research and potential intervention.

Earnings Virality

We examine the determinants and market consequences associated with earnings announcements going viral on social media, a phenomenon we label “earnings virality.” Using a comprehensive panel of historical Twitter data, we find that the typical earnings announcement receives relatively little social media coverage, but others go viral on social media, quickly reaching the feeds of millions of people. We find that viral earnings announcements generally have Twitter content that is more extreme in tone and contains less unique content. Further, earnings virality is positively associated with revenue surprises, investor recognition, retail investor ownership, and retail investor trading around the announcement. Earnings virality appears to be detrimental to markets, as it coincides with lower market liquidity and slower price formation. Overall, our evidence suggests that user-driven dissemination through social media platforms, when amplified and taken to extreme levels, may be harmful to markets.

Detection of Students’ Problems in Distance Education Using Topic Modeling and Machine Learning

Following the rapid spread of COVID-19 to all the world, most countries decided to temporarily close their educational institutions. Consequently, distance education opportunities have been created for education continuity. The abrupt change presented educational challenges and issues. The aim of this study is to investigate the content of Twitter posts to detect the arising topics regarding the challenges of distance education. We focus on students in Saudi Arabia to identify the problems they faced in their distance education experience. We developed a workflow that integrates unsupervised and supervised machine learning techniques in two phases. An unsupervised topic modeling algorithm was applied on a subset of tweets to detect underlying latent themes related to distance education issues. Then, a multi-class supervised machine learning classification technique was carried out in two levels to classify the tweets under discussion to categories and further to sub-categories. We found that 76,737 tweets revealed five underlying themes: educational issues, social issues, technological issues, health issues, and attitude and ethical issues. This study presents an automated methodology that identifies underlying themes in Twitter content with a minimum human involvement. The results of this work suggest that the proposed model could be utilized for collecting and analyzing social media data to provide insights into students’ educational experience.

Twitter content curation as an antidote to hybrid warfare during Russia’s invasion of Ukraine

Concern regarding information disorders has been magnified by the proliferation of social networks. Since its occupation of Crimea in 2014, Russia has been spewing disinformation both inside and outside its borders, giving rise to a hybrid conflict, which since 24 February 2022 has become an invasion. Faced with this flood of malicious information on social networks, fact-checkers assume the role of content curators, relying on contextualization, verification, and literacy improvement to reduce such noise. This work studies the Twitter activity of three Spanish fact-checkers (Newtral, EFE Verifica, and Maldito Bulo), to fight this new epidemic of disinformation. The sample (n = 397) was subjected to content analysis to study the evolution of the verifications and their reaction capacity, the purpose of their activity, the formats in which the content is presented, and their distribution and interaction as revealed by reactions on Twitter. The results reveal a rapid, albeit repetitive, response of the fact-checkers to the invasion, support from them to end the internationalization of hoaxes, a reliance on denials and contextualization rather than literacy improvement, unattractive formats, and a distribution and impact that demonstrate a greater reaction to sensational and emotive content.

Content Analysis of the Main Twitter Accounts of Saudi Universities

With unemployment amongst graduates in Saudi Arabia, social media platforms especially Twitter play a big part in communication. However, do universities utilise Twitter enough when engaging with employers and the community? This study analyses the post content in the main Twitter accounts of 13 private universities and 20 public universities in Saudi Arabia. Automated data extraction via Twitter API was used to gather the data. From the extracted tweets, 350 tweets were selected from private university accounts and 400 tweets from public university accounts to code for content analysis. The post content from both university types was informative and there was no presence of opinions. The most popular category of content for private universities was university events whereas for public universities it was general advertisements. References to employer engagement were more present in private university tweets compared to public university tweets. The study did highlight that more material needs to be posted relating to knowledge and technology transfer.

How does the public discuss gene-editing in agriculture? An analysis of Twitter content

As people form their opinion about gene editing applications in agriculture, they are utilizing social media to seek and share information and opinions on the topic. Understanding how the public discusses this technology will influence the development of effective messaging and practitioner engagement in the conversation. The purpose of this study was to describe the characteristics of Twitter content related to applications of gene editing in agriculture. Social media monitoring facilitated a quantitative, descriptive analysis of public Twitter content related to the topic. A Meltwater social media monitor collected N = 13,189 relevant tweets for analysis, revealing the amount of conversation regarding gene editing in agriculture, the number of contributing Twitter users, and the reach of the conversation which was relatively stable over the life of the study. In contrast, engagement with the topic rose with the sentiment of tweets becoming increasingly positive. News organization accounts had the most reach while a mix of news accounts and personal accounts garnered the greatest engagement. These results demonstrate an opportunity for agricultural and science communicators to create affirmative messaging about gene editing in agriculture delivered through news media Twitter accounts potentially increasing the reach and engagement in the social system and with science communication.

Toward Using Twitter for PrEP-Related Interventions: An Automated Natural Language Processing Pipeline for Identifying Gay or Bisexual Men in the United States.

BackgroundPre-exposure prophylaxis (PrEP) is highly effective at preventing the acquisition of HIV. There is a substantial gap, however, between the number of people in the United States who have indications for PrEP and the number of them who are prescribed PrEP. Although Twitter content has been analyzed as a source of PrEP-related data (eg, barriers), methods have not been developed to enable the use of Twitter as a platform for implementing PrEP-related interventions.ObjectiveMen who have sex with men (MSM) are the population most affected by HIV in the United States. Therefore, the objectives of this study were to (1) develop an automated natural language processing (NLP) pipeline for identifying men in the United States who have reported on Twitter that they are gay, bisexual, or MSM and (2) assess the extent to which they demographically represent MSM in the United States with new HIV diagnoses.MethodsBetween September 2020 and January 2021, we used the Twitter Streaming Application Programming Interface (API) to collect more than 3 million tweets containing keywords that men may include in posts reporting that they are gay, bisexual, or MSM. We deployed handwritten, high-precision regular expressions—designed to filter out noise and identify actual self-reports—on the tweets and their user profile metadata. We identified 10,043 unique users geolocated in the United States and drew upon a validated NLP tool to automatically identify their ages.ResultsBy manually distinguishing true- and false-positive self-reports in the tweets or profiles of 1000 (10%) of the 10,043 users identified by our automated pipeline, we established that our pipeline has a precision of 0.85. Among the 8756 users for which a US state–level geolocation was detected, 5096 (58.2%) were in the 10 states with the highest numbers of new HIV diagnoses. Among the 6240 users for which a county-level geolocation was detected, 4252 (68.1%) were in counties or states considered priority jurisdictions by the Ending the HIV Epidemic initiative. Furthermore, the age distribution of the users reflected that of MSM in the United States with new HIV diagnoses.ConclusionsOur automated NLP pipeline can be used to identify MSM in the United States who may be at risk of acquiring HIV, laying the groundwork for using Twitter on a large scale to directly target PrEP-related interventions at this population.

Development of argument based opinion mining model with sentimental data analysis from twitter content

AbstractIn present scenario, social networks have developed massive in practice and society impact. Specifically, micro‐blogging is on trend in various platforms, such as Twitter, Instagram to evaluate public opinions for various issues. In recent times, some methods are developed for evaluating Twitter messages, based on the sentiment and opinions presented in tweets, corresponding to the hash‐tags and keywords. However, these models have some issues in handling the contradictory content and inconsistent data. Considering with this, this article presents an argument based opinion mining model with sentimental data analysis, for extracting specific argument, which is assessed in bottom‐up manner from the content from society emotion's reflects on the messages. Moreover, this model makes the user to pull out the arguments from a document set, which contains content from commercial sites, to extract the mostly argued positive and negative content. This model use natural language processing techniques, extraction of argument words for defining the decisions. The classification Naive Bayes classification is used for categorizing the results widely under agreed or disagreed. The experimental results prove that the proposed model provides feasible and appropriate results in argument analysis from Twitter content.

SOCIAL MEDIA, YOUTH AND ENVIRONMENTAL LOW-RISK ACTIVISM: A CASE STUDY OF SAVESHARKS INDONESIA CAMPAIGN ON TWITTER

The lack of understanding and awareness on shark’s conservation in Indonesia is still largely found. Evidently, it could be seen from the massive reducing number of shark’s population, irresponsible consumption of shark meat, and constant activities on illegal, unregulated and unreported shark fishing and finning in Indonesia. On the other hand, shark existence is considered critical for sustainability in the marine area. In 2012, Savesharks Indonesia has started an online campaign on Twitter as their effort to raise public awareness of the critical role of sharks’ conservation. However, social media activism has the possibility to lead into ‘slacktivism’, or ‘feel good’ activism; without creating a meaningful impact for the cause. This article is intended to explore the relationship between online environmental activism and its implication toward the cause. On the first stage, we analyse the Twitter content in 2014 Savesharks campaign – based on Lovejoy & Saxton (2012) framework of communicative functions in Twitter environmental advocacy. On the second stage, we conduct interviews with informants who represent Savesharks’ campaign audience – to investigate the level of activism that the audience refers to in this campaign. Results indicate that Savesharks Indonesia online activism has reached the action stage where the audience were involved not only in online realm of activism but also in offline forms. However, the online activism of Savesharks Indonesia was reflecting more of a “populist political activism” (Lim, 2013). As such, it is still questionable whether the activism could lead to tangible political actions (e.g. influencing the environmental policies).
 

Constructing twitter corpus of Iraqi Arabic Dialect (CIAD) for sentiment analysis

The number of Twitter users in Iraq has increased significantly in recent years. Major events, the political situation in the country, had a significant impact on the content of Twitter and affected the tweets of Iraqi users. Creating an Iraqi Arabic Dialect corpus is crucial for sentiment analysis to study such behaviors. Since no such corpus existed, this paper introduces the Corpus of Iraqi Arabic Dialect (CIAD). The corpus has been collected, annotated and made publicly accessible to other researchers for further investigation. Furthermore, the created corpus has been validated using eight different combinations of four feature-selections approaches and two versions of Support Vector Machine (SVM) algorithm. Various performance measures were calculated. The obtained accuracy, 78 %, indicates a promising potential application.

Afrika’da Kuşak ve Yol Girişimi Üzerinde Güven ve Şüpheler: Etiyopya Twitter Uzamındaki Kanaatlerin Tematik İçerik Analizi

China`s Belt and Road Initiative is a massive infrastructural project that Ethiopia is encompassed. Yet, in Ethiopia, public opinion over the subject has never been homogenous as there are both apparent faiths that the initiative would positively contribute to Ethiopia’s economy, and suspicions that it is merely China`s veiled ambition to accelerate its expansion in global economy and politics, intensifying the concerns that China will not be any different from former colonial powers for African nations. Besides mainstream media coverage, much of the debate over this initiative has increasingly happened on social networking sites as attributable to their relative accessibility and autonomy. By employing a thematic content analysis of Twitter contents generated by opinion technicians during the 2019 Belt and Road Initiative Forum in Beijing, this article examines how opinion technicians over the Ethiopian Twittersphere discuss the initiative.

Monitoring event-driven dynamics on Twitter: a case study in Belarus

Analysts of social media differ in their emphasis on the effects of message content versus social network structure. The balance of these factors may change substantially across time. When a major event occurs, initial independent reactions may give way to more social diffusion of interpretations of the event among different communities, including those committed to disinformation. Here, we explore these dynamics through a case study analysis of the Russian-language Twitter content emerging from Belarus before and after its presidential election of August 9, 2020. From these Russian-language tweets, we extracted a set of topics that characterize the social media data and construct networks to represent the sharing of these topics before and after the election. The case study in Belarus reveals how misinformation can be re-invigorated in discourse through the novelty of a major event. More generally, it suggests how audience networks can shift from influentials dispensing information before an event to a de-centralized sharing of information after it.

Linking COVID-19 Perception With Socioeconomic Conditions Using Twitter Data

We, as humans, are constantly in relation with our environment. Sudden changes in our living media may alter the way we perceive ourselves and our environment in various ways. Coronavirus (COVID-19) outbreak is a great example of a sudden change. The outbreak influenced means of work, study, socialization, and communication in unprecedented ways. In our study, we investigate the topic dynamics of Twitter content sharing for the Republic of Turkey. We have analyzed 1.3 million tweets containing the keyword “korona” shared between February 24, 2020 and May 2, 2020. Our study has three key outcomes. The first one is, after the official announcement of first COVID-19 case in Turkey, rate of COVID-19-related content sharing decreases and hygiene-, lifestyle-, and anxiety-related tweets increase. Second, we see that a number of friends and followers influence content sharing dynamics where accounts sharing COVID-19 News-related content have more followers compared to accounts that share the remaining contents. Finally, motivated by the World Bank’s Poverty Monitoring Technical Note, we inquired the effect of income on content sharing and found that GDP per capita of the author’s city is more influential on COVID-19 News-related content sharing compared to the population and COVID-19 cases per 1 00 000 people. The lower the GDP per capita, the higher the COVID-19 News-related content sharing is. Also, our model indicates that lower income and population along with high rates of COVID-19 cases per 1 00 000 people are associated with increased COVID-19 News-related content sharing.

Politicians and Racism on Twitter: Citizen Reactions and New Forms of Social Participation

This paper is a study case that analyses the various functioning logics of racism and anti-racism in Twitter, specifically following the publication of a polemic tweet against immigration by Pablo Casado, president of PP, which aligns itself on the conservative “right” of Spanish politics. The aim is to provide knowledge –still scarce in research on the subject– on the characteristics of racist discourse –and its confrontation– in digital spaces. It is a case study analysing how the selected political discourse, as elite discourse, elicits reactions and provokes social participation in digital spaces. Methodologically, it is based on Twitter content analysis, utilizing both quantitative (frequency of topics) and qualitative (articulation of arguments) approaches. We worked through the NVivo analysis software and selected a sample of tweets, that were then coded and analysed in depth, responding to the politician’s overall message. The results point to the existence of a strong rejection to the politician's words. However, there was an absence of a visibly explicit and significant anti-racism in the ensuing retorts. The support received for his racist comment was in the minority, although some were very aggressive.

109. Exploring The Discussion Of Nexplanon (Etonogestrel Birth Control Implant) On Twitter

The Federal Drug Administration (FDA) oversees approval of medical devices, including birth control implants. Few studies have examined tools to assess the risk of federally approved medical devices already on the market, such as social media. Social media is increasingly used by adolescents and young adults (AYAs) to report health issues. Thus, this study examined Twitter content discussing the medical birth control implant, Nexplanon, including differences in content across time.

Social media monitoring of suicidal content and change in trends of Japanese twitter content around the Zama suicide pact slayings

Social media-related suicides remain a serious issue despite efforts to address this problem. On October 31st, 2017, the Zama Suicide Pact Slayings were reported in which a criminal targeted victims who expressed suicidal ideation using social media. We analyzed how communication on Twitter was used concerning suicidal ideation in relation to the slayings. We extracted data from 1,246 Twitter accounts using the hashtag “#I_want_to_die” between October 1st to November 30th, 2017. We performed thematic content analysis to identify the characteristics of these Twitter accounts and their tweets. The number and categories of related posts from before and after the slayings were compared. Relevant online communication increased from 159 to 1,037 Twitter accounts after the incident. Before the incident, most accounts had tweets related to suicide, mental health issues, or sought to make connections. After the incident, most tweets from these accounts were related to opinions (especially offensive ones) or non-prevention-oriented advertisements. The results suggest that the number of accounts tweeting suicidal-related themes decreased after the homicides, while the number of accounts posting offensive opinions and non-preventive advertisements increased. This implies that the efficacy of social media-based prevention measures may be undermined by this content.