Twitter Language Research Articles

Tweeting environmental pollution: Analyzing twitter language to uncover its correlation with county-level obesity rates in the United States

BackgroundEnvironmental pollution has been linked to obesogenic tendencies. Using environmental-related posts from Twitter (now known as X) from U.S. counties, we aim to uncover the association between Twitter linguistic data and U.S. county-level obesity rates. MethodsAnalyzing nearly 300 thousand tweets from January 2020 to December 2020 across 207 U.S. counties, using an innovative Differential Language Analysis technique and drawing county-level obesity data from the 2020 Food Environment Atlas to identify distinct linguistic features in Twitter relating to environmental-related posts correlated with socioeconomic status (SES) index indicators, obesity rates, and obesity rates controlled for SES index indicators. We also employed predictive modeling to estimate Twitter language's predictive capacity for obesity rates. ResultsResults revealed a negative correlation between environmental-related tweets and obesity rates, both before and after adjusting for SES. Contrarily, non-environmental-related tweets showed a positive association with higher county-level obesity rates, indicating that individuals living in counties with lower obesity rates tend to tweet environmental-related language more frequently than those living in counties with higher obesity rates. The findings suggest that linguistic patterns and expressions employed in discussing environmental-related themes on Twitter can offer unique insights into the prevailing cross-sectional patterns of obesity rates. ConclusionsAlthough Twitter users are a subset of the general population, incorporating environmental-related tweets and county-level obesity rates and using a novel language analysis technique make this study unique. Our results indicated that Twitter users engaging in more active dialog about environmental concerns might exhibit healthier lifestyle practices, contributing to reduced obesity rates.

Gender-inclusive Language in Twitter

Abstract: Gender-inclusive language serves to recognize women and people with non-binary gender identities as part of the primary group, and it has been seen in various languages, including Arabic (Berger 2019), English (Berger 2019; Mathews 1995), Hebrew (Berger 2019), Portuguese (Secretaria de Comunicação 2021), Spanish (Berger 2019; Guzmán Stein 2004), and Swedish (Berger 2019; Gustafsson Sendén et al. 2015), among others (Berger 2019). In Spanish, for example, speakers exchange the generic masculine – o gender morpheme for morphemes like –@, – x , or – e . Through an analysis of the use of these morphemes in the word nosotros by Twitter users in or around Madrid, it is argued that a language change based on analogy to non-gender-marked forms such as estudiante and inteligente is in progress (Campbell 2013; Gubb 2013; Moreno, p.c.; Pesce and Etchezahar 2019). Data are manually extracted from tweets posted during the first through the tenth of each month in the years 2010, 2014, and 2018. The data reveal that no inclusive morphemes were used in 2010, but that by 2018, the most frequent gender-inclusive morpheme was –@, – x was the second-most frequent, and – e was the least frequent. The frequency of these morphemes corresponds with their novelty: – @ was created first, and – e is the most recent. Future studies should examine uses of gender-inclusive language in later years, as well as the demographic characteristics of the Twitter users.

Twitter language samples reflect collective emotional responses following political leaders’ rhetoric during the pandemic across four countries

Twitter language samples reflect collective emotional responses following political leaders’ rhetoric during the pandemic across four countries

BERTuit: Understanding Spanish language in Twitter with transformers

AbstractThe appearance of complex attention‐based language models such as BERT, RoBERTa or GPT‐3 has allowed to address highly complex tasks in a plethora of scenarios. However, when applied to specific domains, these models encounter considerable difficulties. This is the case of Social Networks such as Twitter, an ever‐changing stream of information written with informal and complex language, where each message requires careful evaluation to be understood even by humans given the important role that context plays. Addressing tasks in this domain through Natural Language Processing involves severe challenges. When powerful state‐of‐the‐art multilingual language models are applied to this scenario, language specific nuances get lost in translation. To face these challenges we present BERTuit, the largest transformer proposed so far for Spanish language, pre‐trained on a massive dataset of 230 M Spanish tweets using RoBERTa optimization. Our motivation is to provide a powerful resource to better understand Spanish Twitter and to be used on applications focused on this social network, with special emphasis on solutions devoted to tackle the spreading of misinformation in this platform. BERTuit is evaluated on several tasks and compared against M‐BERT, XLM‐RoBERTa and XLM‐T, very competitive multilingual transformers. The utility of our approach is shown with applications, in this case: an unsupervised methodology to visualize groups of hoaxes; and supervised profiling of authors spreading disinformation.

Linguistic Methodologies to Surveil the Leading Causes of Mortality: Scoping Review of Twitter for Public Health Data.

Twitter has become a dominant source of public health data and a widely used method to investigate and understand public health-related issues internationally. By leveraging big data methodologies to mine Twitter for health-related data at the individual and community levels, scientists can use the data as a rapid and less expensive source for both epidemiological surveillance and studies on human behavior. However, limited reviews have focused on novel applications of language analyses that examine human health and behavior and the surveillance of several emerging diseases, chronic conditions, and risky behaviors. The primary focus of this scoping review was to provide a comprehensive overview of relevant studies that have used Twitter as a data source in public health research to analyze users' tweets to identify and understand physical and mental health conditions and remotely monitor the leading causes of mortality related to emerging disease epidemics, chronic diseases, and risk behaviors. A literature search strategy following the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) extended guidelines for scoping reviews was used to search specific keywords on Twitter and public health on 5 databases: Web of Science, PubMed, CINAHL, PsycINFO, and Google Scholar. We reviewed the literature comprising peer-reviewed empirical research articles that included original research published in English-language journals between 2008 and 2021. Key information on Twitter data being leveraged for analyzing user language to study physical and mental health and public health surveillance was extracted. A total of 38 articles that focused primarily on Twitter as a data source met the inclusion criteria for review. In total, two themes emerged from the literature: (1) language analysis to identify health threats and physical and mental health understandings about people and societies and (2) public health surveillance related to leading causes of mortality, primarily representing 3 categories (ie, respiratory infections, cardiovascular disease, and COVID-19). The findings suggest that Twitter language data can be mined to detect mental health conditions, disease surveillance, and death rates; identify heart-related content; show how health-related information is shared and discussed; and provide access to users' opinions and feelings. Twitter analysis shows promise in the field of public health communication and surveillance. It may be essential to use Twitter to supplement more conventional public health surveillance approaches. Twitter can potentially fortify researchers' ability to collect data in a timely way and improve the early identification of potential health threats. Twitter can also help identify subtle signals in language for understanding physical and mental health conditions.

Predicting U.S. county opioid poisoning mortality from multi-modal social media and psychological self-report data

Opioid poisoning mortality is a substantial public health crisis in the United States, with opioids involved in approximately 75% of the nearly 1 million drug related deaths since 1999. Research suggests that the epidemic is driven by both over-prescribing and social and psychological determinants such as economic stability, hopelessness, and isolation. Hindering this research is a lack of measurements of these social and psychological constructs at fine-grained spatial and temporal resolutions. To address this issue, we use a multi-modal data set consisting of natural language from Twitter, psychometric self-reports of depression and well-being, and traditional area-based measures of socio-demographics and health-related risk factors. Unlike previous work using social media data, we do not rely on opioid or substance related keywords to track community poisonings. Instead, we leverage a large, open vocabulary of thousands of words in order to fully characterize communities suffering from opioid poisoning, using a sample of 1.5 billion tweets from 6 million U.S. county mapped Twitter users. Results show that Twitter language predicted opioid poisoning mortality better than factors relating to socio-demographics, access to healthcare, physical pain, and psychological well-being. Additionally, risk factors revealed by the Twitter language analysis included negative emotions, discussions of long work hours, and boredom, whereas protective factors included resilience, travel/leisure, and positive emotions, dovetailing with results from the psychometric self-report data. The results show that natural language from public social media can be used as a surveillance tool for both predicting community opioid poisonings and understanding the dynamic social and psychological nature of the epidemic.

Terrorist Attacks, Cultural Incidents, and the Vote for Radical Parties: Analyzing Text from Twitter

AbstractWe study the role of perceived threats from other cultures induced by terrorist attacks and criminal events on public discourse and support for radical‐right parties. We develop a rule which allocates Twitter users to electoral districts in Germany and use a machine‐learning method to compute measures of textual similarity between the tweets they produce and tweets by accounts of the main German parties. Using the exogenous timing of attacks, we find that, after an event, Twitter language becomes on average more similar to that of the main radical‐right party, AfD. The result is driven by a larger share of tweets discussing immigrants and Muslims, common AfD topics, and by a more negative sentiment of these tweets. Shifts in language similarity are correlated with changes in vote shares between federal elections. These results point to the role of perceived threats from minorities on the success of nationalist parties.

Filling in the white space: Spatial interpolation with Gaussian processes and social media data

Full national coverage below the state level is difficult to attain through survey-based data collection. Even the largest survey-based data collections, such as the CDC's Behavioral Risk Factor Surveillance System or the Gallup-Healthways Well-being Index (both with more than 300,000 responses p.a.) only allow for the estimation of annual averages for about 260 out of roughly U.S. 3,000 counties when a threshold of 300 responses per county is used. Using a relatively high threshold of 300 responses gives substantially higher convergent validity–higher correlations with health variables–than lower thresholds but covers a reduced and biased sample of the population. We present principled methods to interpolate spatial estimates and show that including large-scale geotagged social media data can increase interpolation accuracy. In this work, we focus on Gallup-reported life satisfaction, a widely-used measure of subjective well-being. We use Gaussian Processes (GP), a formal Bayesian model, to interpolate life satisfaction, which we optimally combine with estimates from low-count data. We interpolate over several spaces (geographic and socioeconomic) and extend these evaluations to the space created by variables encoding language frequencies of approximately 6 million geotagged Twitter users. We find that Twitter language use can serve as a rough aggregate measure of socioeconomic and cultural similarity, and improves upon estimates derived from a wide variety of socioeconomic, demographic, and geographic similarity measures. We show that applying Gaussian Processes to the limited Gallup data allows us to generate estimates for a much larger number of counties while maintaining the same level of convergent validity with external criteria (i.e., N = 1,133 vs. 2,954 counties). This work suggests that spatial coverage of psychological variables can be reliably extended through Bayesian techniques while maintaining out-of-sample prediction accuracy and that Twitter language adds important information about cultural similarity over and above traditional socio-demographic and geographic similarity measures. Finally, to facilitate the adoption of these methods, we have also open-sourced an online tool that researchers can freely use to interpolate their data across geographies.

World Business Leaders’ Performativity of Twitter Language as Social Action

This large longitudinal qualitative study advances the theory of leadership language in the digital era by investigating the power of performativity and social action in softening the bust and constituting the boom of the world economy. More than a million tweets were collected from Twitter accounts of 70 CEOs or/and Chairmen of major global companies who had a total number of tweets above hundred and a total number of followers between 10,000 and 50,000,000 on their Twitter. This represents the entire collection of CEOs tweets between March 2006 (when Twitter was launched) and August 2019. We used those tweets to extract over 8.5 million linguistic tokens to analyze the frequency of words and hashtags, networks of bigrams frequency, and sentiments. Our findings show a possibility that the performativity of those leaders’ language on social media contributes to taking the world economy out of one of the toughest economic recessions to a record-long economic expansion built on social media influence and influencers’ impact.

Detection of Hate Speech and Offensive Language in Twitter Using Sentiment Analysis

Abstract: The dramatic development of online media, for example, Twitter and local area gatherings has upset correspondence and content distributing, but at the same time is progressively misused for the spread of disdain discourse and the association of disdain based exercises. The secrecy and portability managed by such media has made the rearing and spread of disdain discourse – in the long run prompting disdain wrongdoing – easy in a virtual land scape past the domains of conventional law requirement. Existing techniques in the identification of disdain discourse principally cast the issue as a regulated report grouping task [33]. These can be partitioned into two classifications: one depends on manual element designing that are then devoured by calculations, for example, SVM, Naive Bayes, and Logistic Regression [3, 9, 11, 15, 19, 23, 35–39] (exemplary techniques); the other addresses the later profound learning worldview that utilizes neural organizations to consequently learn multi-facets of dynamic highlights from crude information [13, 26, 30, 34] (profound learning strategies). In this technique We show that it is a significantly more testing task, as our examination of the language in the commonplace datasets shows that disdain discourse needs interesting, discriminative highlights and hence is found in the 'long tail' in a dataset that is hard to find. We then, at that point propose Deep Neural Network structures filling in as highlight extractors that are especially powerful for catching the semantics of disdain discourse. Our techniques are assessed on the biggest assortment of disdain discourse datasets dependent on Twitter, and are demonstrated to have the option to beat best in class by up to 6 rate focuses in large scale normal F1, or 9 rate focuses in the seriously difficult instance of recognizing derisive substance. As an intermediary to evaluate and think about the semantic attributes of disdain and non-disdain Tweets, we additionally propose to contemplate the 'uniqueness' of the jargon for each class. Keywords: Classic Methods; DNN; Detection of hate speech and offensive language in Twitter; Sentimental Analysis

Dude, srsly?: The Surprisingly Formal Nature of Twitter's Language

Twitter has become the de facto information sharing and communication platform. Given the factors that influence language on Twitter - size limitation as well as communication and content-sharing mechanisms - there is a continuing debate about the position of Twitter's language in the spectrum of language on various established mediums. These include SMS and chat on the one hand (size limitations) and email (communication), blogs and newspapers (content sharing) on the other. To provide a way of determining this, we propose a computational framework that offers insights into the linguistic style of all these mediums. Our framework consists of two parts. The first part builds upon a set of linguistic features to quantify the language of a given medium. The second part introduces a flexible factorization framework, soclin, which conducts a psycholinguistic analysis of a given medium with the help of an external cognitive and affective knowledge base. Applying this analytical framework to various corpora from several major mediums, we gather statistics in order to compare the linguistics of Twitter with these other mediums via a quantitative comparative study. We present several key insights: (1) Twitter's language is surprisingly more conservative, and less informal than SMS and online chat; (2) Twitter users appear to be developing linguistically unique styles; (3) Twitter's usage of temporal references is similar to SMS and chat; and (4) Twitter has less variation of affect than other more formal mediums. The language of Twitter can thus be seen as a projection of a more formal register into a size-restricted space.

"How Old Do You Think I Am?" A Study of Language and Age in Twitter

In this paper we focus on the connection between age and language use, exploring age prediction of Twitter users based on their tweets. We discuss the construction of a fine-grained annotation effort to assign ages and life stages to Twitter users. Using this dataset, we explore age prediction in three different ways: classifying users into age categories, by life stages, and predicting their exact age. We find that an automatic system achieves better performance than humans on these tasks and that both humans and the automatic systems have difficulties predicting the age of older people. Moreover, we present a detailed analysis of variables that change with age. We find strong patterns of change, and that most changes occur at young ages.

Monitoring Depression Trends on Twitter During the COVID-19 Pandemic: Observational Study.

BackgroundThe COVID-19 pandemic has affected people’s daily lives and has caused economic loss worldwide. Anecdotal evidence suggests that the pandemic has increased depression levels among the population. However, systematic studies of depression detection and monitoring during the pandemic are lacking.ObjectiveThis study aims to develop a method to create a large-scale depression user data set in an automatic fashion so that the method is scalable and can be adapted to future events; verify the effectiveness of transformer-based deep learning language models in identifying depression users from their everyday language; examine psychological text features’ importance when used in depression classification; and, finally, use the model for monitoring the fluctuation of depression levels of different groups as the disease propagates.MethodsTo study this subject, we designed an effective regular expression-based search method and created the largest English Twitter depression data set containing 2575 distinct identified users with depression and their past tweets. To examine the effect of depression on people’s Twitter language, we trained three transformer-based depression classification models on the data set, evaluated their performance with progressively increased training sizes, and compared the model’s tweet chunk-level and user-level performances. Furthermore, inspired by psychological studies, we created a fusion classifier that combines deep learning model scores with psychological text features and users’ demographic information, and investigated these features’ relations to depression signals. Finally, we demonstrated our model’s capability of monitoring both group-level and population-level depression trends by presenting two of its applications during the COVID-19 pandemic.ResultsOur fusion model demonstrated an accuracy of 78.9% on a test set containing 446 people, half of which were identified as having depression. Conscientiousness, neuroticism, appearance of first person pronouns, talking about biological processes such as eat and sleep, talking about power, and exhibiting sadness were shown to be important features in depression classification. Further, when used for monitoring the depression trend, our model showed that depressive users, in general, responded to the pandemic later than the control group based on their tweets (n=500). It was also shown that three US states—New York, California, and Florida—shared a similar depression trend as the whole US population (n=9050). When compared to New York and California, people in Florida demonstrated a substantially lower level of depression.ConclusionsThis study proposes an efficient method that can be used to analyze the depression level of different groups of people on Twitter. We hope this study can raise awareness among researchers and the public of COVID-19’s impact on people’s mental health. The noninvasive monitoring system can also be readily adapted to other big events besides COVID-19 and can be useful during future outbreaks.

Partisan Differences in Twitter Language Among US Legislators During the COVID-19 Pandemic: Cross-sectional Study

BackgroundAs policy makers continue to shape the national and local responses to the COVID-19 pandemic, the information they choose to share and how they frame their content provide key insights into the public and health care systems.ObjectiveWe examined the language used by the members of the US House and Senate during the first 10 months of the COVID-19 pandemic and measured content and sentiment based on the tweets that they shared.MethodsWe used Quorum (Quorum Analytics Inc) to access more than 300,000 tweets posted by US legislators from January 1 to October 10, 2020. We used differential language analyses to compare the content and sentiment of tweets posted by legislators based on their party affiliation.ResultsWe found that health care–related themes in Democratic legislators’ tweets focused on racial disparities in care (odds ratio [OR] 2.24, 95% CI 2.22-2.27; P<.001), health care and insurance (OR 1.74, 95% CI 1.7-1.77; P<.001), COVID-19 testing (OR 1.15, 95% CI 1.12-1.19; P<.001), and public health guidelines (OR 1.25, 95% CI 1.22-1.29; P<.001). The dominant themes in the Republican legislators’ discourse included vaccine development (OR 1.51, 95% CI 1.47-1.55; P<.001) and hospital resources and equipment (OR 1.22, 95% CI 1.18-1.25). Nonhealth care–related topics associated with a Democratic affiliation included protections for essential workers (OR 1.55, 95% CI 1.52-1.59), the 2020 election and voting (OR 1.31, 95% CI 1.27-1.35), unemployment and housing (OR 1.27, 95% CI 1.24-1.31), crime and racism (OR 1.22, 95% CI 1.18-1.26), public town halls (OR 1.2, 95% CI 1.16-1.23), the Trump Administration (OR 1.22, 95% CI 1.19-1.26), immigration (OR 1.16, 95% CI 1.12-1.19), and the loss of life (OR 1.38, 95% CI 1.35-1.42). The themes associated with the Republican affiliation included China (OR 1.89, 95% CI 1.85-1.92), small business assistance (OR 1.27, 95% CI 1.23-1.3), congressional relief bills (OR 1.23, 95% CI 1.2-1.27), press briefings (OR 1.22, 95% CI 1.19-1.26), and economic recovery (OR 1.2, 95% CI 1.16-1.23).ConclusionsDivergent language use on social media corresponds to the partisan divide in the first several months of the course of the COVID-19 public health crisis.

Well-Being Depends on Social Comparison: Hierarchical Models of Twitter Language Suggest That Richer Neighbors Make You Less Happy.

Psychological research has shown that subjective well-being is sensitive to social comparison effects; individuals report decreased happiness when their neighbors earn more than they do. In this work, we use Twitter language to estimate the well-being of users, and model both individual and neighborhood income using hierarchical modeling across counties in the United States (US). We show that language-based estimates from a sample of 5.8 million Twitter users replicate results obtained from large-scale well-being surveys - relatively richer neighbors leads to lower well-being, even when controlling for absolute income. Furthermore, predicting individual-level happiness using hierarchical models (i.e., individuals within their communities) out-predicts standard baselines. We also explore language associated with relative income differences and find that individuals with lower income than their community tend to swear (f*ck, sh*t, b*tch), express anger (pissed, bullsh*t, wtf), hesitation (don't, anymore, idk, confused) and acts of social deviance (weed, blunt, drunk). These results suggest that social comparison robustly affects reported well-being, and that Twitter language analyses can be used to both measure these effects and shed light on their underlying psychological dynamics.

Actuallyin Nordic tweets

‘Native-like’ use of discourse markers is a good indicator of language proficiency. Analysing four subcorpora of English-language tweets posted by Twitter users from the Nordic countries of Finland, Norway, and Sweden, this study considers the effects of discursive context and L1 influence on the correlation between semantic function and sentence position of the discourse marker actually. The study shows that both predictors appear to have a significant effect. A more formal context predicts more standard punctuation, distribution of the pragmatic functions, and placement of the discourse marker, and L1 influence is reflected in the preferred sentence position, with a substantial and significant difference observed between the Finnic and Germanic L1s. Furthermore, the study shows that while the discourse marker actually is significantly more frequent in colloquial Twitter language than in spoken English, the frequency is significantly lower and in line with spoken English in more constrained contexts.

Detecting and Classifying Toxic Language in Twitter using Machine Learning

Today international on-line content material has turned out to be a first-rate part due to growth in the use of net. Individuals of various societies and instructive foundation can speak through this platform. Therefore, for automatic detection of poisonous content, we need to distinguish between hate speech and offensive language. Here a way to robotically stumble on and classify tweets on Twitter into 3 commands: hateful, offensive and easy is proposed. We do not forget n-grams as functions and by way of passing their time period frequency-inverse document frequency (TFIDF) values to numerous system gaining knowledge of fashions using Twitter dataset, we perform comparative evaluation of the models. We work towards classification and comparison of different classifiers using the combination of best feature from each type of feature extraction and determining which model works best for the purpose of classification of tweets into hate-speech, offensive language or neither.

The Combination constraint and Semantic function of ‘-dab-’, ‘-lob-’, ‘-seuleob-’ - Focusing on the Twitter language -

The purpose of this study is to reveal the combination constraint and semantic function of the adjective derivational suffixes, ‘-dab-’, ‘-lob-’, ‘-seuleob-’ based on Twitter language. To this end, it collected the derivatives of ‘-dab-’, ‘-lob-’, ‘-seuleob-’ and analyzed the aspects of use in Tweets written from 1 to 31 Oct. 2018. These suffixes have taken similar meanings but have been divided into different phonological, syntactic and semantic combination constraints. On Twitter, however, the suffix perceived as losing productivity in the past was also being actively used since the constraints were greatly weakened. The weakening of the combination constraint occurred in the syntactic and semantic levels, and, in particular, as the weakening of the semantic constraint was common all of ‘-dab-’, ‘-lob-’, ‘-seuleob-’, it showed that their boundaries of meaning were being broken down. Only it showed the differences in the way that ‘-dab-’ strongly reveals the equality between preceding base and subject, ‘-lob-’ functions to abstract the concrete base, and ‘-seuleob-’ shows to come close to nature of base while maintaining the substance of base somewhat.

Authority (de)legitimation in the border wall Twitter discourse of President Trump

Abstract The construction of a wall along the US/Mexico border was one of the main political platforms upon which the 2016 US presidential election campaign was fought. Ahead of the upcoming 2020 US presidential election, and with the border wall still not yet built or funded, this article uses the authorisation component of Van Leeuwen’s (2007) framework for the discourse of legitimation to show how President Donald J. Trump has sought legitimacy for the construction of the border wall. Data is taken from Trump’s @realDonaldTrump Twitter postings between October 18th, 2018 and February 3rd, 2019, a period inclusive of the longest federal government shutdown in US history. We show how Trump’s Twitter language is frequently accompanied by evidence-less attacks on sources of rival opinion or information, while the president tends to reaffirm himself as the exclusive source of credible and truthful information.

An Analysis of Abbreviation in Twitter Status of Hollywood Pop Singers

Abbreviation is the process of cutting one word, some parts of the word and the combination of words into words in the new form. Abbreviation is used by all kinds of people who share situations of interests and it is often used in informal situation. The use of abbreviation is not only found in daily conversation, discussion, or speech, but also found in online interaction in social media. This study aims to find out the types and the meaning of abbreviation words and types internet language in Twitter used by Hollywood pop singers. This research used Schendl (2001) and Weakly (2006) theories about five types of abbreviation words. This study is descriptive qualitative research. The data were taken from tweets that have been posted by Hollywood pop singers on June 2019. The result showed that from 47 data that have been collected, there were four types out of abbreviation that have been found from five types of Schendl an Weakly abbreviations. They are clipping (14,9%), initialism (34,0%), blend (10,6%), contraction (40,4%) and acronym is not found from the data. From the finding it can be concluded that the type of abbreviation which is dominantly used by the singers in Twitter is contraction.