Twitter Comments Research Articles

Discursive Practices during an Electoral Cycle: Public Opinion and Political Disillusionment on Twitter

Electoral cycles are highly charged, politically intense moments that influence public discourse. Political elites, citizens, and the traditional news media seek to generate opinions and interactions on social networks. This research is motivated by the following questions: What are the communicative and deliberative practices used in these spaces? Is it possible to identify the characteristics that –through deliberative conversations– potentially foster or undermine democratic debate, particularly when using populist and polarizing discourse? Using a mixed methods approach, we apply a social network analysis tool to track conversations and identify the volume of political discourse (N=346,000). Using selective and staged filtering, we identified posts from nine Peruvian Twitter accounts during the September 2022 electoral campaign in Lima: three candidates, three media outlets, and three accounts with high levels of engagement. Our data comprised the comments from these nine accounts and was extracted using an Application Programming Interface (API). Subsequently, we carried out ethnographic content analysis on publications with more than 30 comments. In this phase, we analyzed Twitter comments using a codebook to identify deliberative practices and user responses. Our findings underscore the significant role of principal media outlets in shaping political conversation on Twitter. We also discovered that attempts to interact and deliberate were often overshadowed by heated comments attempting to impose opinions on others. Most importantly, our research reveals a widespread disillusionment with politics, political institutions, and even the political preferences of fellow citizens in Peru, highlighting a key challenge for democratic discourse on social media.

Part of the Job? The Effect of Exposure to the Online Intimidation of Politicians on Political Ambition

It is increasingly common for politicians to experience online harassment. Even though numerous studies show the prevalence and impact of online abuse towards politicians, the broader societal impact of this phenomenon is understudied. This study focuses on the deterring effect that witnessing online abuse towards politicians could have on bystanders, testing to what extent online intimidation influences the appeal of political office. This pre-registered project examines whether exposure to abusive or sexist Twitter comments towards politicians weakens the interest of UK citizens, women especially, in pursuing political office. We hypothesized that seeing intimidating Twitter comments addressed to politicians leads to less political ambition in bystanders, especially if their gender identity aligns with that of the targeted politician. Sexist abusive comments were expected to have a greater impact on women participants than non-gendered abusive comments. The findings show that citizens are generally not easily deterred by exposure to abuse. However, we find differential effects that are concerning for political participation and female representation. There is evidence that sexist comments do discourage women observers. Respondents also generally espouse very negative expectations about the online experience of political aspirants.

The Performance Analysis of Graph Neural Network (GNN) and Convolutional Neural Network (CNN) Algorithms for Cyberbullying Detection in Twitter Comments

Cyberbullying incidents have surged due to the expansion of social media network and advancements in internet technology, presenting a substantial challenge in online communities. Previous studies employing Support Vector Machine (SVM) techniques have exhibited promising outcomes, achieving a superior accuracy of 71.25%. However, recognizing the dynamic nature of cyberbullying behaviors and the necessity for more robust detection methodologies, this research explores cyberbullying detection on Twitter utilizing Convolutional Neural Network (CNN) and Graph Neural Network (GNN). The selection of CNN and GNN is motivated by the deficiencies observed in prior SVM-based approaches and the capacity of neural network to capture intricate patterns in textual and network data. The GNN consistently outperforms CNN in terms of F1 score, accuracy, precision, and recall. With only 20 epochs, GNN achieves an accuracy of 80.25%, surpassing CNN's 68.43%. Through GNN optimization, its accuracy reaches 89.04% after 100 epochs, underscoring its efficacy in Twitter cyberbullying detection.

Sentiment Analysis about Electric Motorbikes in Indonesia Using Twitter Data

Along with the rapid development of technology, various types of transportation have experienced increased innovation in shapes, colours,models, and even engines. However, one thing that needs special attention is the number of pollutants or emissions released by vehicles. One effort to reduce emissions is increasing the production of Battery-Based Electric Motorized Vehicles. Battery-based electric vehicles developed in Indonesia include electric cars and electric motorcycles. Among these types of electric vehicles, Indonesian society widely adopts electric motorcycles. However, sales of electric motorbikes were only 15 thousand units, lower than sales of petrol motorbikes which reached 5 million units. This study contributes to understanding further how the community responds to electric motorcycles in detail through sentiment analysis on social media data. Consumer acceptance of electric motorcycles can be seen from the numerous active Twitter users in Indonesia who provide positive and negative comments on the presence of electric motorcycles. Text information based on public comments in Indonesia via Twitter is collected using Sentiment Analysis in R Studio. Twitter comments will be classified into positive, negative, and neutral groups. The results show 63% positive, 21% unfavourable, and 14% neutral opinions. This condition means that Indonesian society accepts and has a supportive opinion of the presence of electric motorcycles. The government and entrepreneurs can use this information to create electric motorcycles that align with the community's preferences.

Engagement of stakeholders in the development of generic medical and nursing curricula in Tanzania through media outlets and social media

Background. The Catholic University of Health and Allied Health Sciences, Kilimanjaro Christian Medical University College and Muhimbili University of Health and Allied Sciences teamed up to address gaps in the training curricula for medical doctors and nurses with the aim of developing harmonised curricular templates that would be comprehensive and suit national objectives. Objectives. To share experiences and lessons learned in engaging stakeholders using social media as an additional method to collect information on curricular gaps and ideas to develop harmonised medical and nursing curricula. Methods. Pictures, newsprint extracts, and videos of face-to-face events in the curriculum development process and project and a curriculum advert were posted on Facebook, Twitter and YouTube and used to engage stakeholders, including regulatory bodies, health professional councils, graduates, students, practitioners, health training institutions, private and public employers, internship supervisors, faculty and the public. Reactions, comments and post insights from stakeholder postings, in both Swahili and English, were analysed manually and categorised into four types of messages: congratulatory, comments on the process and expected outcomes, curricular gaps, and faculty development needs to implement new curricula. Results. A total of 69 290 stakeholders were engaged via Facebook and 229 via Twitter. A total of 13 553 (19.6%) Facebook and 179 (78.2%) Twitter comments were directly related to gaps in current curricula and graduate competencies. Other inputs received through social media were on faculty development needs to implement harmonised curricula and the project in general. Conclusion. This was the first attempt to engage curriculum stakeholders using social media for the development of harmonised curricula in Tanzania. The impact of social media in providing relevant inputs for curriculum development was significant. Use of social media, with multiple language options, is an economical and efficient way to reach a large number of stakeholders for curriculum quality improvement.

Tesla Stock Prediction Based on Twitter Comments’ Sentiment Feature Extraction and Ensemble Networks

As people become more aware of environmental issues, Tesla, a pillar of the global electric vehicle market, has become a hot commodity in recent years. The prediction of Tesla’s share price is also a hot topic in the investment market. This experiment extracts sentiment factors of tweets about Tesla’s comments, combined with Tesla’s historical stock price as the dataset for training, testing and inspection. This experiment builds time series prediction models based on LSTM, XGBoost and RF algorithms to predict Tesla’s stock price, and evaluates the prediction effectiveness of the three algorithms based on the fit and error of the prediction results. The analysis of the data shows that XGBoost has the best fit and the lowest error among the three algorithms, and that the sentiment factor has its unique utility as raw data. The experimental results also empirically demonstrate the applicability of sentiment factor analysis and the three algorithms LSTM, XGBoost and RF in the field of stock price prediction.

Word embedding for detecting cyberbullying based on recurrent neural networks

<span lang="EN-US">The phenomenon of cyberbullying has spread and has become one of the biggest problems facing users of social media sites and generated significant adverse effects on society and the victim in particular. Finding appropriate solutions to detect and reduce cyberbullying has become necessary to mitigate its negative impacts on society and the victim. Twitter comments on two datasets are used to detect cyberbullying, the first dataset was the Arabic cyberbullying dataset, and the second was the English cyberbullying dataset. Three different pre-trained global vectors (GloVe) corpora with different dimensions were used on the original and preprocessed datasets to represent the words. Recurrent neural networks (RNN), long short-term memory (LSTM), Bidirectional LSTM (BiLSTM), gated recurrent unit (GRU), and Bidirectional GRU (BiGRU) classifiers utilized, evaluated and compared. The GRU outperform other classifiers on both datasets; its accuracy on the Arabic cyberbullying dataset using the Arabic GloVe corpus of dimension equal to 256D is 87.83%, while the accuracy on the English datasets using 100 D pre-trained GloVe corpus is 93.38%.</span>

UJARAN KEBENCIAN NETIZEN INDONESIA PADA AKUN TWITTER ES TEH: TINJAUAN LINGUISTIK FORENSIK

This study aims to describe the form of netizens' hate speech on the @esteh_indonesia Twitter account which can be categorized as a violation of the law. This research is included in the qualitative research used to describe netizens' hate speech in @esteh_indonesia Twitter comments using forensic linguistic reviews. The methods and techniques used are the listening method and the note-taking technique. Listening is done by observing the hate speech of netizens on the Twitter account @esteh_indonesia. After that, the data that has been selected purposively is copied and then analyzed by collecting data that has been recorded in one folder, classifying and analyzing the data based on the form of hate speech with the guidelines of the Police circular regarding hate speech on social media. The results of the study prove that there are five forms of netizens' hate speech in the Twitter comments of the @esteh.indonesia account: 1) insults, 2) defamation, 3) blasphemy, 4) provoking, and 5) hoax. The results of this study can increase public awareness about the existence of hate speech on social media platforms, especially Twitter. This can encourage people to be more careful in interacting online.

Dynamical features and traveling wave structures of the perturbed Fokas-Lenells equation in nonlinear optical fibers

In this study, the new complex wave solutions of the perturbed Fokas-Lenells (p-FL) equation, which has applications in nonlinear optical fibers are obtained using a new extended direct algebraic method. This model represents recent electronic communications like Internet blogs, facebook communication and twitter comments. The obtained solutions are the different classes of traveling wave structures with singular solutions Type-I & II, dark-singular, dark, and dark-bright solutions. Furthermore, stability conditions for the computed structures are reported. Also, graphical representations of some particular structures are shown by taking the specific values of the constants. The ordinary differential equation (ODE) obtained from a traveling wave transformation is converted into a dynamical system using Galilean transformation. The phase plane analysis is done for different values of the controlled parameters d 1 and d 3. A perturbation term is added to analyze the chaotic dynamics, and plots indicate that the system shows the chaotic dynamics. Also, sensitivity analysis shows that the system is sensitive to initial conditions. The conclusion is accounted for toward the end.

Sentiment Analysis of Twitter Comments Using Naive Bayes Classifier

Social media has a significant role in how people express their emotions and elaborate on their opinions in today's culture. There are many new forms of social media, and Twitter is one of them. In this experiment, the sentiment of pre-processed Twitter comment data was examined using naive Bayes and logistic regression techniques. In order to categorise the emotional tendency of text for Twitter comments, a naive Bayesian classifier is created. In processing this material, the Naive Bayes and logistic regression models' benefits and drawbacks are compared and summarised. Naive Bayes can achieve good accuracy with binary emotion analysis. The accuracy of the naive Bayes model is 0.06 points higher than that of logic training under identical processing settings, and the recall rate is 0.05 points higher.

What do Twitter comments tell about news article bias? Assessing the impact of news article bias on its perception on Twitter

News stories circulating online, especially on social media platforms, are nowadays a primary source of information. Given the nature of social media, news no longer are just news, but they are embedded in the conversations of users interacting with them. This is particularly relevant for inaccurate information or even outright misinformation because user interaction has a crucial impact on whether information is uncritically disseminated or not. Biased coverage has been shown to affect personal decision-making. Still, it remains an open question whether users are aware of the biased reporting they encounter and how they react to it. The latter is particularly relevant given that user reactions help contextualize reporting for other users and can thus help mitigate but may also exacerbate the impact of biased media coverage.This paper approaches the question from a measurement point of view, examining whether reactions to news articles on Twitter can serve as bias indicators, i.e., whether how users comment on a given article relates to its actual level of bias. We first give an overview of research on media bias before discussing key concepts related to how individuals engage with online content, focusing on the sentiment (or valance) of comments and on outright hate speech. We then present the first dataset connecting reliable human-made media bias classifications of news articles with the reactions these articles received on Twitter. We call our dataset BAT - Bias And Twitter. BAT covers 2,800 (bias-rated) news articles from 255 English-speaking news outlets. Additionally, BAT includes 175,807 comments and retweets referring to the articles.Based on BAT, we conduct a multi-feature analysis to identify comment characteristics and analyze whether Twitter reactions correlate with an article’s bias. First, we fine-tune and apply two XLNet-based classifiers for hate speech detection and sentiment analysis. Second, we relate the results of the classifiers to the article bias annotations within a multi-level regression. The results show that Twitter reactions to an article indicate its bias, and vice-versa. With a regression coefficient of 0.703 (p<0.01), we specifically present evidence that Twitter reactions to biased articles are significantly more hateful. Our analysis shows that the news outlet’s individual stance reinforces the hate-bias relationship. In future work, we will extend the dataset and analysis, including additional concepts related to media bias.

Hate Speech Detection in Indonesia Twitter Comments Using Convolutional Neural Network (CNN) and FastText Word Embedding

Hate Speech Detection in Indonesia Twitter Comments Using Convolutional Neural Network (CNN) and FastText Word Embedding

Analisis Sentimen Tanggapan Masyarakat Terhadap Calon Presiden 2024 Ridwan Kamil Menggunakan Metode Naive Bayes Classifier

Reaction to public facts about the election of the presidential candidate Ridwan Kamil, which will later be obtained, the data is taken from Twitter based on these problems, it is necessary to do sentiment analysis research. Based on the results of this study, the classification process for the Naïve Bayes Classifier has 3 scenarios for dividing training data and test data, namely with 90%:10% training data, the test data produces an accuary value of 85.43%, a recall value of 100.00%, and a precision of 85.33%. For training data 80%: 20% of the test data produces an accuracy value of 86.38%, a recall of 100.00% and a precision value of 86.38% and for data on the distribution of training data 70%: 30% of the test data produces an accuary value of 84.29 %, 100.00% recall and 84.29% precision. From the tweet data that has been used, there are 1262 positive comments and 242 negative comments. These results prove that the Naïve Bayes classifier is very good for conducting sentiment analysis on Twitter comments about the 2024 presidential candidate Ridwan Kamil. The naïve Bayes classifier process gets the highest accuracy value of 86.38% by dividing the training data 80%:20% test data.

#TreatmentResistantDepression: A qualitative content analysis of Tweets about difficult-to-treat depression.

Treatment-resistant depression (TRD) is depression unresponsive to antidepressants and affects 55% of British primary care users with depression. Current evidence is from secondary care, but long referral times mean general practitioners (GPs) manage TRD. Studies show that people with depression use Twitter to form community and document symptoms. However, Twitter remains a largely unexplored space of documented patient experience. Twitter data could provide valuable insights into learning about primary care experiences of TRD. In this study, we explored Twitter comments and conversations about TRD and produced patient-driven recommendations. Tweets from UK-based users were collected manually and using a browser extension in June 2021. Conventional content analysis was used to provide an overview of the Tweets, followed by interpretation to understand why Twitter may be important to people with TRD. A total of 415 Tweets were organised into five clusters: self-diagnosis, symptoms, support, small wins and condition experts. These Tweets were interpreted as showing Twitter as a community for people with TRD. People had a collective sense of illness identity and were united in their experiences of TRD. However, users in the community also highlighted the absence of effective GP care, leading users to position themselves as condition experts. Users shared advice from a place of lived experience with the community but also shared potentially harmful information, including recommendations about nonevidence-based medications. Findings illuminate the benefits of the TRD Twitter community and also highlight that the perception of a lack of knowledge and support from GPs may lead community members to advise nonevidenced-based medications. This study was led by a person with lived experience of TRD and bipolar. Two public contributors with mental health conditions gave feedback on our study protocol and results.

Identifying factors influencing COVID-19 vaccine uptake in Finland - a qualitative study using social media data.

Vaccine demand creation requires understanding what is driving the uptake of the vaccine. 24 Qualitative research methods are paramount to gaining a localized understanding of behavioral 25 drivers and barriers to vaccine uptake, but they are often underutilized. This is a qualitative study that 26 used public comments on the Facebook and Twitter posts of the Finnish Institute for Health and 27 Welfare (THL) as data sources to identify behavioral drivers for COVID-19 vaccine uptake in 28 Finland. The participatory data analysis utilized thematic analysis and the Theoretical Domains 29 Framework (TDF). NVIVO was used to assist in the coding process. The greatest number of FB and 30 Twitter comments were linked with six TDF domains: knowledge, environmental context and 31 resources, beliefs in consequences, beliefs in capabilities, social and professional role, and social 32 influences. The domains included 15 themes that were interlinked. The knowledge domain 33 overlapped with all other domains. By using public discourse on Facebook and Twitter, and rapid 34 qualitative data analysis methods within a behavioral insight framework, this study adds to the 35 emerging knowledge about behavioral drivers of COVID-19 vaccines that can be used by public 36 health experts to enhance the uptake of vaccines during future pandemics and epidemics.

Cross-Border Tensions Affecting Sports

&#x0D; &#x0D; &#x0D; This study aimed to explore cross-border tensions affecting sports and how people reacted to the decision made in the wake of those tensions. Fifty Twitter comments were sampled using a purposive method to gain knowledge about the themes in reactions to bold decisions made by Ramiz Raja. Results revealed 11 main themes. The literature suggests that disputes are brought up by politicians, which reflects the general population. The results show that the public does not hold the same views as the politicians, and people are more accepting of moving past conflicts. People are very protective of cricket in both nations. Current events have shown that despite politicians working hard to keep nations apart, the people do not completely agree.&#x0D; &#x0D; &#x0D;

Trump's Misogynistic Linguistic Elements and His Rhetorical Device

This paper analyzes the misogynistic linguistic elements of Donald Trump, from the first GOP debate (6 Aug. 2015) on a FOX channel and a CNN interview (7 Aug. 2015) concerning Fox News anchor and the debate moderator Megyn Kelly. His Twitter comments and live interview comments concerning women were projected to show his misogyny. If we consider that Trump became the American Republican presidential candidate at that time, these misogynistic linguistic elements seemed to be of little importance to Trump's supporters.&#x0D; Trump projected a smart image in terms of his linguistic strategy; justifying himself through the use of membership categorization, rhetorical use of mitigation and lexical cohesion such as anaphoric, exophoric references, and repetition as a rhetorical device. In his remarks, he showed a hierarchical structure on the importance of misogyny. For him and his supporters, the misogynistic characteristics seemed to be less important than other important issues facing the United States. Despite misogynistic controversies in Trump’s speeches, he succeeded in his path toward the election.

Machine Learning Based Sentiment Analysis of Message on Twitter

In the age of information explosion, it is a very important and challenging task to extract the required features from a huge amount of information. The emotion expressed by the text information is one of the most important features of the information. However, there is not much research in this field, so it is of great significance and exploratory to the text sentence emotion analysis. In order to compare and explore better feasibility, both the sequential neural network model and the random forest model were built. Through the contrast between the two models, the machine learning of datasets composed of Twitter comments is carried out to analyze the emotion of the text, and emotion extraction is the research topic of this paper. In this paper, emotion analysis was studied in the order of data processing, model building, and result analysis, and an accuracy of about 90% was finally achieved, which is a good result, from which it can be seen that the constructed neural network model plays a good role.

As metáforas no discurso de/sobre André Ventura: as metáforas que constrói André Ventura e as que constroem sobre ele

Through an interdisciplinary approach between Cognitive Linguistics and Critical Discourse Analysis, we aim to identify and analyse conceptual and situational metaphors present in the interventions of MP André Ventura in the Portuguese Parliament and in tweets both from the MP’s official account and from other accounts that publish comments about him. For this purpose, we collected data from 15 reports of the Assembly of the Republic Diary and from 98 tweets, then proceeded to manually annotate the metaphors and to search them by target domain. The results show many conceptual metaphors in their interventions, where they function as an attack on the political opponent, and of situated metaphors in Twitter comments, discursively and cognitively constructed by network users based on cultural assumptions. It was also concluded that metaphori-cal rhetoric in populist discourse represents a powerful discursive strategy, allowing power plays in political interactions, discursive manipulations and triggering emotions.

Sentiment Analysis on Moroccan Dialect based on ML and Social Media Content Detection

As technology continues to evolve, humans tend to follow suit, and currently social media has taken place as the defacto method of communication. As it tends to happen with verbal communication, people express their opinions in written form and through an analysis of their words, one can extract what an individual wants from a product, a topic, or an event. By looking at the emotions expressed in such content, governments, businesses, and people can learn a lot that can help them improve their strategies. Therefore, in this study, we will use different algorithms to improve the Moroccan sentiment classification. The first step is to gather and prepare Moroccan Dialectal Arabic Twitter comments. Then, a lot of different combinations of extraction (n-grams) and weighting schemes (BOW/ TF-IDF) and word embedding for feature construction are applied to get the best classification models. We used Naive Bayes, Random Forests, Support Vector Machines, and Logistic regression and LSTM to classify the data we prepared. Our machine learning approach, which incorporates sentiment analysis, was designed to analyze Twitter comments written in Modern Standard Arabic or Moroccan Dialectal Arabic. As a final benchmark of our paper, we were simply a sliver shy away from the 70% mark in our accuracy by relying on the SVM algorithm. Although not a game-changing result, this was enough to encourage us to continue developing our model further.