Tweet Messages Research Articles

Occupation Prediction with Multimodal Learning from Tweet Messages and Google Street View Images

Occupation Prediction with Multimodal Learning from Tweet Messages and Google Street View Images

Semiotic Violence

Semiotic Violence

SENTIMENT ANALYSIS BASED ON MACHINE LEARNING METHODS ON TWITTER DATA USING oneAPI

With the gradual development of technology, the use of digital social networks has becomewidespread. As a result of this increased use, the information sharing process has accelerated and increased.This has created an enormous amount of data on digital social networks. With the increasing amount ofdata, it has made it possible to make inferences about people, products, companies and many other areas.Various fields of study have emerged to process and analyze the data. One of these fields of study issentiment analysis. Sentiment analysis is the analysis of data from different sources such as text, sound orimage by classifying the attitude towards a subject as positive, negative or neutral. In this study, sentimentanalysis was performed using Sentiment140 dataset created through Twitter application, one of the digitalsocial network platforms. After preprocessing on the dataset containing English Tweet messages, BernoulliNaive Bayes, Linear Support Vector Machine, Logistic Regression and artificial neural network algorithmsLSTM and CNN methods were used in a hybrid way. Classification results were evaluated with f1 score.Accuracy rates of 0.80%, 0.82%, 83%, 85% were achieved for Bernoulli Naive Bayes, Linear SupportVector Machine, Logistic Regression, LSTM and CNN classifications respectively.

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

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

Threat Modelling and Detection Using Semantic Network for Improving Social Media Safety

Social media provides a free space to users to post their information, opinions, feelings, etc. Also, it allows users to easily and simultaneously communicate with each other. As a result, threat detection in social media is critical for ensuring the user’s safety and preventing suspicious activities such as criminal behavior, hate speech, ethnic conflicts and terrorist plots. These suspicious activities have a negative impact on the community’s life and cause tension and social unrest among individuals in both inside and outside of cyberspace. Furthermore, with the recent popularity of social networking sites, the number of discussions containing threats is increasing, causing fear in various parties, whether at the individual or state level. Moreover, these social networking service providers do not have complete control over the content that users post. In this paper, we propose to design a threat detection model on Twitter using a semantic network. To achieve this aim, we designed a threat semantic network, named, ThrNet that will be integrated in our proposed threat detection model called, DetThr. We compared the performance of our model (DetThr) with a set of well-known Machine Learning algorithms. Results show that the DetThr model achieves an accuracy of 76% better than Machine Learning algorithms. It works well with an error rate of forecasting threatening tweet messages as non-threatening (false negatives) is about 29%, while the error rate of forecasting non-threatening tweet messages as threatening (false positives) is about 19%.

E-Cigarette–Related Health Beliefs Expressed on Twitter Within the U.S.

This mixed-methods study analyzed English-language U.S.-based Twitter posts related to E-cigarette use from February 2021. Posts were manually identified as health-related or not and, if health-related, whether they were posted by an E-cigarette user. A random selection of 1,000 health-related tweets from 986 unique E-cigarette users were qualitatively content analyzed for theory of planned behavior constructs as well as nature and tone of each tweet message. Using quantitative semantic network analysis, relationships among the identified topics and sentiment-specific conversation patterns were explored. The most salient health-related conversation topics of E-cigarette users, health beliefs corresponding to each theory of planned behavior construct, and major motivational contexts of E-cigarette use were identified. Seven topics emerged in positive tweets: smoking cessation, social impact generation, controls over addiction, therapeutic effects on physical and mental health, social support, device attachment, and peer influence. Nine topics emerged in negative tweets: side effects on physical health, vaping addiction, lack of E-cigarette regulations, peer pressure, increased risk of COVID-19, side effects on mental health, no help in smoking cessation, social conflict, and polysubstance use. Most assertions for E-cigarette benefits were not substantiated. Jokes in tweets appeared to contribute to the view of vaping as an attractive, enjoyable, safe, and fun activity. Discussions about positive aspects of E-cigarette use were concentrated on a few related topics, whereas tweets discouraging E-cigarette use presented a diverse, less related set of topics. The results provide insights into the drivers of E-cigarette use behaviors. E-cigarette user perspectives gathered from social media may inform research to guide future prevention and cessation interventions.

APPLICATION OF LEXICON BASED FOR SENTIMENT ANALYSIS OF COVID-19 BOOSTER VACCINATIONS ON TWITTER SOCIAL MEDIA USING NAÏVE BAYES METHOD

To combat the Covid-19 epidemic, the government issues laws governing vaccination implementation. Health Minister Number Ten of 2021 issued the regulation. This program raises advantages and disadvantages, necessitating examination through feedback. The opinions and narratives that individuals share on social media sites like Twitter can be used to get feedback. This work seeks to construct a model to assess public opinion of the Covid-19 Booster Vaccination by using the Lexicon Based technique to identify sentiment on tweet data. Naïve Bayes and logistic regression are the classification techniques employed in this study. The comparison of the two methods' findings reveals that Logistic Regression, with an accuracy of 72%, is superior to Naïve Bayes, which has an accuracy of 70%. There were 607 tweet messages from Twitter that were processed. From January 1 to July 30, 2022, the model was tested for its ability to interpret public opinion on Twitter. The model found that people's attitudes toward the COVID-19 booster shot tended to be favorable. It can be developed by including datasets for additional research. For further research, it can be developed by adding datasets.

Social media analysis of car parking behavior using similarity based clustering

This paper investigates car parking users’ behaviors from social media perspective using social network based analysis of online communities revealed by mining the associated hashtags in Twitter. We propose a new interpretable community detection approach for mapping user’s car parking behavior by combining Clique, K-core and Girvan–Newman community detection algorithms together with a content-based analysis that exploits polarity, relative frequency and dominant topics. Twitter API was used to collect relevant data by tracking popular car-parking hashtags. A social network graph is constructed using a similarity-based analysis. Finally, interpretable communities are inferred by monitoring the outcomes of clique, K-core and Girvan–Newman community detection algorithms. This interpretability is linked to the aggregation of keywords, hashtags and/or location attributes of the tweet messages as well as a visualization module that enables interaction with users. In parallel, a global trend analysis investigates parking types and Twitter influence with respect to both sentiment polarity and dominant trends (extracted using KeyBERT based approach) is performed. The implementation of this social media analytics has uncovered several aspects associated to car-parking behaviors. A comparison with some state-of-the-art community detection methods has also been carried out and revealed some similarities with our developed approach.

The OPM Data Breach: An Investigation of Shared Emotional Reactions on Twitter

This paper investigates the shared emotional responses of Twitter users in the aftermath of a massive data breach, a crisis event known as the Office of Personnel Management (OPM) data breach of 2015. This breach impacted the lives of several million individuals due to the exposure of sensitive and personally identifying information. We take a data exploration approach to analyzing over 18,000 tweet messages of the ensuing discussion that took place after public notification that the breach had occurred. The resulting analysis reveals that although the emotions of anxiety, anger, and sadness may initially appear erratic, at an aggregate level, the public display of these emotions corresponds to the situational awareness of the breach event. Further, our analysis finds that this relationship extends to the sharing of emotions, indicating that those participating in the conversation congregate around a sense of shared emotional experience. Finally, an in-depth analysis of the ensuing dialogue identifies the most salient conversational drivers of these emotions, revealing breach concepts most significantly related to each emotion. Based on the results, we present propositions that draw from this analysis to inform emotional response characteristics that emerge over the duration of such crisis events. The results of this study can inform organizational practices and policy making in the context of response to crisis events such as data breaches.

Analyzing Public Concerns Over COVID-19 Variants Using Social Media

SARS-CoV-2, or more popularly known COVID-19 has claimed more than 5.5 million lives since it has been declared as a global pandemic. Similar to other viruses, COVID-19 is also undergoing several mutations and has many variants like Alpha, Beta, Gamma, Delta, Omicron and others. With so many variants, social media users are confused and posting their frustrations and angers with Tweets or Posts in public social media platforms. These publicly accessible social media posts provide a wealth of information for a social scientist or political leader or a strategic decision maker. This study demonstrates a feasible approach to extract meaningful critical information from social media posts. By programmatically accessing Twitter database from 11th January 2022 till 20th January 2022, we retrieved almost 9 K Tweet messages on 6 different keywords like “COVID Variants”, “Omicron”, “Alpha Variant”, “Beta Variant”, “Gamma Variant” and “Delta Variant”. Results were compared against metrics like users, posts, engagement, and influence. Omicron was found to be the most popular topic compared to other variants with an influence score of 70.2 million and 2.1 K posts during the monitored period. The most popular sources for influences on COVID-19 Variant related posts were found to be @reuters with 24.2M, @forbes with 17.4M, @timesofindia with 14.2M and @inquirerdotnet with 3.4 followers. This study also found out that the most popular Tweet languages were English followed by French and Dutch. Lastly, this study ranked user mentions, word frequency (with word cloud) and hashtags for COVID-19 Variant related twitter posts during the monitored timeframe.

Veracity assessment by single and multi-source identification algorithms during the crisis

Social networks have become a popular communication tool for information sharing. Twitter offers access to data and provides a significant opportunity to analyze data. During pandemics, Twitter becomes a big source for the dispersal of unverified information. In social media, it is difficult to find the sources of rumors. To tackle this problem the authors have developed a hybrid rumor centrality algorithm for rumor source detection in social networks. The authors propose an S-RSI algorithm for identifying a single rumor centre and an M-RSI algorithm for identifying the propagations of multiple rumor centres in the thread of conversation. The proposed rumor centrality algorithm efficiently predicts the rumor disseminating possibilities in a conversation tree with the aid of graph theoretical approach. The authors have evaluated the performance of the algorithms on the PHEME dataset containing seven real-time event conversational trees based on the tweet messages. The results show that the proposed is best suitable in finding the rumor source centre with a high probability in social media during a crisis.

Harnessing Artificial Intelligence for Health Message Generation: The Folic Acid Message Engine.

BackgroundCommunication campaigns using social media can raise public awareness; however, they are difficult to sustain. A barrier is the need to generate and constantly post novel but on-topic messages, which creates a resource-intensive bottleneck.ObjectiveIn this study, we aim to harness the latest advances in artificial intelligence (AI) to build a pilot system that can generate many candidate messages, which could be used for a campaign to suggest novel, on-topic candidate messages. The issue of folic acid, a B-vitamin that helps prevent major birth defects, serves as an example; however, the system can work with other issues that could benefit from higher levels of public awareness.MethodsWe used the Generative Pretrained Transformer-2 architecture, a machine learning model trained on a large natural language corpus, and fine-tuned it using a data set of autodownloaded tweets about #folicacid. The fine-tuned model was then used as a message engine, that is, to create new messages about this topic. We conducted a web-based study to gauge how human raters evaluate AI-generated tweet messages compared with original, human-crafted messages.ResultsWe found that the Folic Acid Message Engine can easily create several hundreds of new messages that appear natural to humans. Web-based raters evaluated the clarity and quality of a human-curated sample of AI-generated messages as on par with human-generated ones. Overall, these results showed that it is feasible to use such a message engine to suggest messages for web-based campaigns that focus on promoting awareness.ConclusionsThe message engine can serve as a starting point for more sophisticated AI-guided message creation systems for health communication. Beyond the practical potential of such systems for campaigns in the age of social media, they also hold great scientific potential for the quantitative analysis of message characteristics that promote successful communication. We discuss future developments and obvious ethical challenges that need to be addressed as AI technologies for health persuasion enter the stage.

Social media analytics and reachability evaluation - #Diabetes

Background and aimsDiabetes as a lifestyle disorder could be effectively managed by creating awareness among people through social media. Understanding the content of Twitter messages will aid in strategizing health communication about diabetes to the community through Twitter. This study aimed to analyze the content, sentiment, and reachability of diabetes related tweets posted in India. MethodsDiabetes related messages from India were collected via Twitter's Application Programming Interface for April 2019. Themes and subthemes of tweet content were identified from randomly selected tweets. The tweets were coded as the source, themes, and subthemes manually. Sentiment analysis of the tweets was done by a lexicon-based approach. The reachability of tweets was assessed based on re-tweet and favorite counts. ResultsOut of 1840 tweets, 57.28% were from organizations and 42.72% were from individuals. The largest proportion of tweet messages were informative (50.76%), followed by promotional tweets (21.52%). The largest proportion of tweets were positive (40.4%) followed by neutral (31.14%) tweets. Among the six major themes, the diabetes story had the highest reachability. ConclusionsThe outcome of this study would aid public health professionals in planning information dissemination and communication regarding diabetes on Twitter so that the right information reaches a wider population.

Sentiment Analysis Covid-19 Vaccination on Twitter Social Media Using Naïve Bayes Method

The government regulations regarding the implementation of vaccinations to tackle the COVID-19 pandemic. The regulation was issued by the Minister of Health Number ten of 2021. This program raises pros and cons so that it requires feedback for evaluation. Feedback can be obtained from opinions and stories that users convey through social media such as Twitter. This study aims to develop a model to determine public sentiment towards Covid-19 vaccination in three topics, namely the vaccination program, the effect of vaccination and the Covid-19 vaccine. The classification method used in this research is Bernoulli Naïve Bayes and Logistic Regression. The results of the comparison of the two methods show that Bernoulli Naïve Bayes gets better accuracy results. The number of tweet messages processed from Twitter is 5877. The model was tested to read public sentiment on Twitter from 7 September to 21 September 2021. The model concluded that public opinion regarding the vaccination program and the effect of vaccination tended to be positive. And opinions regarding the Covid-19 vaccine topic tend to be neutral. For further research, it can be developed by adding datasets.

소셜미디어 브랜드 메시지의 언어적 특성이 고객인게이지먼트에 미치는 영향

Social media is regarded as a means to increase customer engagement, and the use of social media to induce customer engagement is expanding as a major marketing trend of companies in recent years. According to the speech act theory, it is highly likely that not only the content is trying to convey, but also the communication method affects customer engagement. In this context, this study classified the combinations of various linguistic styles contained in the Twitter messages of global brands into classes, and analyzed the relationship between these classes and customer engagement. For empirical analysis, 17,621 tweet messages from six representative global brands were collected. And after analyzing linguistic characteristics using LIWC 2015, an automated text analysis program, we classified the linguistic style combinations of global brand tweet messages through the Latent Class Analysis method. Using Zero-Inflated Negative Binomial regression analysis, the relationship between the class and customer engagement classified through linguistic style combination was analyzed. As results, the linguistic style combinations of global brand tweet messages were divided into 4 classes, and it was found that there was a difference in customer engagement performance according to each class.

A comprehensive social media data processing and analytics architecture by using big data platforms: a case study of twitter flood-risk messages.

The main objective of the article is to propose an advanced architecture and workflow based on Apache Hadoop and Apache Spark big data platforms. The primary purpose of the presented architecture is collecting, storing, processing, and analysing intensive data from social media streams. This paper presents how the proposed architecture and data workflow can be applied to analyse Tweets with a specific flood topic. The secondary objective, trying to describe the flood alert situation by using only Tweet messages and exploring the informative potential of such data is demonstrated as well. The predictive machine learning approach based on Bayes Theorem was utilized to classify flood and no flood messages. For this study, approximately 100,000 Twitter messages were processed and analysed. Messages were related to the flooding domain and collected over a period of 5 days (14 May – 18 May 2018). Spark application was developed to run data processing commands automatically and to generate the appropriate output data. Results confirmed the advantages of many well-known features of Spark and Hadoop in social media data processing. It was noted that such technologies are prepared to deal with social media data streams, but there are still challenges that one has to take into account. Based on the flood tweet analysis, it was observed that Twitter messages with some considerations are informative enough to be used to estimate general flood alert situations in particular regions. Text analysis techniques proved that Twitter messages contain valuable flood-spatial information.

Examining Tweet Content and Engagement of Canadian Public Health Agencies and Decision Makers During COVID-19: Mixed Methods Analysis.

BackgroundEffective communication during a health crisis can ease public concerns and promote the adoption of important risk-mitigating behaviors. Public health agencies and leaders have served as the primary communicators of information related to COVID-19, and a key part of their public outreach has taken place on social media platforms.ObjectiveThis study examined the content and engagement of COVID-19 tweets authored by Canadian public health agencies and decision makers. We propose ways for public health accounts to adjust their tweeting practices during public health crises to improve risk communication and maximize engagement.MethodsWe retrieved data from tweets by Canadian public health agencies and decision makers from January 1, 2020, to June 30, 2020. The Twitter accounts were categorized as belonging to either a public health agency, regional or local health department, provincial health authority, medical health officer, or minister of health. We analyzed trends in COVID-19 tweet engagement and conducted a content analysis on a stratified random sample of 485 tweets to examine the message functions and risk communication strategies used by each account type.ResultsWe analyzed 32,737 tweets authored by 118 Canadian public health Twitter accounts, of which 6982 tweets were related to COVID-19. Medical health officers authored the largest percentage of COVID-19–related tweets (n=1337, 35%) relative to their total number of tweets and averaged the highest number of retweets per COVID-19 tweet (112 retweets per tweet). Public health agencies had the highest frequency of daily tweets about COVID-19 throughout the study period. Compared to tweets containing media and user mentions, hashtags and URLs were used in tweets more frequently by all account types, appearing in 69% (n=4798 tweets) and 68% (n=4781 tweets) of COVID-19–related tweets, respectively. Tweets containing hashtags also received the highest average retweets (47 retweets per tweet). Our content analysis revealed that of the three tweet message functions analyzed (information, action, community), tweets providing information were the most commonly used across most account types, constituting 39% (n=181) of all tweets; however, tweets promoting actions from users received higher than average retweets (55 retweets per tweet). When examining tweets that received one or more retweet (n=359), the difference between mean retweets across the message functions was statistically significant (P<.001). The risk communication strategies that we examined were not widely used by any account type, appearing in only 262 out of 485 tweets. However, when these strategies were used, these tweets received more retweets compared to tweets that did not use any risk communication strategies (P<.001) (61 retweets versus 13 retweets on average).ConclusionsPublic health agencies and decision makers should examine what messaging best meets the needs of their Twitter audiences to maximize sharing of their communications. Public health accounts that do not currently employ risk communication strategies in their tweets may be missing an important opportunity to engage with users about the mitigation of health risks related to COVID-19.

Would you please like my tweet?! An artificially intelligent, generative probabilistic, and econometric based system design for popularity-driven tweet content generation

An understudied area in the field of social media research is the design of decision support systems that can aid the manager by way of automated message component generation. Recent advances in this form of artificial intelligence has been suggested to allow content creators and managers to transcend their tasks from creation towards editing, thus overcoming a common problem: the tyranny of the blank screen. In this research, we address this topic by proposing a novel system design that will suggest engagement-driven message features as well as automatically generate critical and fully written unique Tweet message components for the goal of maximizing the probability of relatively high engagement levels. Our multi-methods design relies on the use of econometrics, machine learning, and Bayesian statistics, all of which are widely used in the emerging fields of Business and Marketing Analytics. Our system design is intended to analyze Tweet messages for the purpose of generating the most critical components and structure of Tweets. We propose econometric models to judge the quality of written Tweets by way of engagement-level prediction, as well as a generative probability model for the auto-generation of Tweet messages. Testing of our design demonstrates the need to take into account the contextual, semantic, and syntactic features of messages, while controlling for individual user characteristics, so that generated Tweet components and structure maximizes the potential engagement levels.

Ready for the World? Measuring the (Trans-)National Quality of Political Issue Publics on Twitter

This article presents a multi-method research design for measuring the (trans-)national quality of issue publics on Twitter. Online communication is widely perceived as having the potential to overcome nationally bound public spheres. Social media, in particular, are seen as platforms and drivers of transnational communication through which users can easily connect across borders. Transnational interactivity can be expected in particular for policy fields of global concern and elite or activist communication as practiced on Twitter. Nevertheless, there is still a lot of evidence for the enduring national structuration of political communication and publics as it results from a shared language (mostly), culturally defined media markets, established routines of social and political communication, and sociocultural stocks of knowledge. The study goes beyond measuring user interaction and also includes indicators of cross-referential cohesion. It applies a set of computational methods in network and discourse analysis and presents empirical evidence for Twitter communication on climate change being a prime issue of global concern and a globalized policy agenda. For empirical analysis, the study relies on a large Twitter dataset (N ≈ 6m tweets) with tweet messages and metadata collected between 2015 and 2018. Based on basic measurements such as geolocation and language use, the metrics allowed measurement of cross-national user interactions, user centrality in communicative networks, linking behaviour, and hashtag co-occurrences. The findings of the exploratory study suggest that a combined perspective on indicators of user interaction and cross-referential cohesion helps to develop a better and more nuanced understanding of online issue publics.

Automatic 3D Modeling and Reconstruction of Cultural Heritage Sites from Twitter Images

This paper presents an approach for leveraging the abundance of images posted on social media like Twitter for large scale 3D reconstruction of cultural heritage landmarks. Twitter allows users to post short messages, including photos, describing a plethora of activities or events, e.g., tweets are used by travelers on vacation, capturing images from various cultural heritage assets. As such, a great number of images are available online, able to drive a successful 3D reconstruction process. However, reconstruction of any asset, based on images mined from Twitter, presents several challenges. There are three main steps that have to be considered: (i) tweets’ content identification, (ii) image retrieval and filtering, and (iii) 3D reconstruction. The proposed approach first extracts key events from unstructured tweet messages and then identifies cultural activities and landmarks. The second stage is the application of a content-based filtering method so that only a small but representative portion of cultural images are selected to support fast 3D reconstruction. The proposed methods are experimentally evaluated using real-world data and comparisons verify the effectiveness of the proposed scheme.