Microblogging Platform Research Articles

Application of Machine Learning Techniques to Classify Twitter Sentiments Using Vectorization Techniques

The advancements in social networking have empowered open expression on micro-blogging platforms like Twitter. Traditional Twitter Sentiment Analysis (TSA) faces challenges due to rule-based or dictionary algorithms, dealing with feature selection, ambiguity, sparse data, and language variations. This study proposed a classification framework for Twitter sentiment data using word count vectorization and machine learning techniques to reduce the difficulties faced with annotated sentiment-labelled tweets. Various classifiers (Naïve Bayes (NB), Decision Tree (DT), K-Nearest Neighbors (KNN), Logistic Regression (LR), and Random Forest (RF)) were evaluated based on accuracy, precision, recall, F1-score, and specificity. Random Forest outperformed the others with an Area under Curve (AUC) value of 0.96 and an average precision (AP) score of 0.96 in sentiment classification, especially effective with minimal Twitter-specific features.

Scalable intimacy: Rethinking intimate sociality and its problem of scale in digital China

With the development of the Internet in China, people have increasingly turned to digital communication and networking tools for personal connections and social relations. From messaging and dating apps to microblogging and live-streaming platforms, mediated forms of sociality constitute a significant part of people’s intimate lives. Scholarly works on modern and contemporary China have shown how practices and politics of intimacy are formulated and contested against the backdrop of the tensions between the residual culture of Confucianism, the socialist ideological regime, and China’s participation in global capitalism. On the contrary, research on global network societies has foregrounded the role of information and media technologies in mediating intimacy with conceptual frameworks such as programmed sociality and networked intimacy. Bridging these two areas of research, this essay attempts to understand how intimate sociality is configured within China’s burgeoning digital economy and culture. It first provides a review of previous works on forms of intimacy and sociality mediated by network technologies ranging from mobile phones to online dating. Using the popularity of live streaming in current-day China as a case study, this essay then introduces the problem of “scale” to consider how intimacy is mediated at scale with Internet-based media technologies and discuss how scalability shapes intimate lives in Chinese society as a shared rationale behind modernization, network technologies, and new forms of economies. This essay proposes the concept of scalable intimacy to move beyond the platform-centered approach and develop a more contextualized framework for understanding intimate sociality in digital China.

How public opinion of food safety affects green food purchase intentions: The mediating role of insecurity and the moderating role of green label trust

Widespread public opinion of food safety (POFS) has exacerbated consumer concerns about food safety and health, then driving consumers to purchase green food. However, the mechanisms at play in such concerns to consumer behaviour towards green food consumption have not explored, especially in the environment of social media. To fill up this gap, this research adopted a sequential two-stage mixed-methods approach to identify and examine the effects and mechanisms of public opinion of food safety on consumers' green food purchase intentions. In the initial qualitative research phase, the public comments on the microblogging platform was analyzed and we identified four top hot topics: public opinion of food safety, insecurity, green label trust (GLT) and green food purchase intentions (GFPI) based on text mining. In the quantitative phase of the survey-based research, we introduced the protection motivation theory and the hyper-attention cognitive model to construct a conceptual model. Based on this conceptual model, we conducted an empirical test with 1087 online questionnaires. The result showed that POFS had a significant positive effect on consumers' GFPI; consumer insecurity mediated the relationship between POFS and GFPI, and GLT played a moderating role between them. This study illustrates a psychological mechanism by which public opinion of food safety motivates consumers to switch to green food and provides new insights into green food consumption behaviour. It is an essential reference for encouraging green food consumption and promoting sustainability in the food industry.

Transformer-based models for combating rumours on microblogging platforms: a review

The remarkable success of Transformer-based embeddings in natural language tasks has sparked interest among researchers in applying them to classify rumours on social media, particularly microblogging platforms. Unlike traditional word embedding methods, Transformers excel at capturing a word’s contextual meaning by considering words from both the left and right of a word, resulting in superior text representations ideal for tasks like rumour detection on microblogging platforms. This survey aims to provide a thorough and well-organized overview and analysis of existing research on implementing Transformer-based models for rumour detection on microblogging platforms. The scope of this study is to offer a comprehensive understanding of this topic by systematically examining and organizing the existing literature. We start by discussing the fundamental reasons and significance of automating rumour detection on microblogging platforms. Emphasizing the critical role of text embedding in converting textual data into numerical representations, we review current approaches to implement Transformer models for rumour detection on microblogging platforms. Furthermore, we present a novel taxonomy that covers a wide array of techniques and approaches employed in the deployment of Transformer-based models for identifying misinformation on microblogging platforms. Additionally, we highlight the challenges associated with this field and propose potential avenues for future research. Drawing insights from the surveyed articles, we anticipate that promising results will continue to emerge as the challenges outlined in this study are addressed. We hope that our efforts will stimulate further interest in harnessing the capabilities of Transformer models to combat the spread of rumours on microblogging platforms.

An efficient method for disaster tweets classification using gradient-based optimized convolutional neural networks with BERT embeddings

Event of the disastrous scenarios are actively discussed on microblogging platforms like Twitter which can lead to chaotic situations. In the era of machine learning and deep learning, these chaotic situations can be effectively controlled by developing efficient methods and models that can assist in classifying real and fake tweets. In this research article, an efficient method named BERT Embedding based CNN model with RMSProp Optimizer is proposed to effectively classify the tweets related disastrous scenario. Tweet classification is carried out via some of the popular the machine learning algorithms such as logistic regression and decision tree classifiers. Noting the low accuracy of machine learning models, Convolutional Neural Network (CNN) based deep learning model is selected as the primary classification method. CNNs performance is improved via optimization of the parameters with gradient based optimizers. To further elevate accuracy and to capture contextual semantics from the text data, BERT embeddings are included in the proposed model. The performance of proposed method - BERT Embedding based CNN model with RMSProp Optimizer achieved an F1 score of 0.80 and an Accuracy of 0.83. The methodology presented in this research article is comprised of the following key contributions:•Identification of suitable text classification model that can effectively capture complex patterns when dealing with large vocabularies or nuanced language structures in disaster management scenarios.•The method explores the gradient based optimization techniques such as Adam Optimizer, Stochastic Gradient Descent (SGD) Optimizer, AdaGrad, and RMSprop Optimizer to identify the most appropriate optimizer that meets the characteristics of the dataset and the CNN model architecture.•“BERT Embedding based CNN model with RMSProp Optimizer” – a method to classify the disaster tweets and capture semantic representations by leveraging BERT embeddings with appropriate feature selection is presented and models are validated with appropriate comparative analysis.

Quantifying polarization in online political discourse

In an era of increasing political polarization, its analysis becomes crucial for the understanding of democratic dynamics. This paper presents a comprehensive research on measuring political polarization on X (Twitter) during election cycles in Spain, from 2011 to 2019. A wide comparative analysis is performed on algorithms used to identify and measure polarization or controversy on microblogging platforms. This analysis is specifically tailored towards publications made by official political party accounts during pre-campaign, campaign, election day, and the week post-election. Guided by the findings of this comparative evaluation, we propose a novel algorithm better suited to capture polarization in the context of political events, which is validated with real data. As a consequence, our research contributes a significant advancement in the field of political science, social network analysis, and overall computational social science, by providing a realistic method to capture polarization from online political discourse.

The Koo Dataset: An Indian Microblogging Platform with Global Ambitions

The Koo Dataset: An Indian Microblogging Platform with Global Ambitions

Evaluation of adjective and adverb types for effective Twitter sentiment classification.

Twitter, the largest microblogging platform, has reported more than 330 million active users in recent years. Many users express their sentiments about politics, sports, products, personalities, etc. Sentiment analysis has emerged as a specialized branch of machine learning in which tweets are binary-classified to provide sentimental insights. A major step in sentiment classification is feature selection, which primarily revolves around parts of speech (POS). Few techniques merely focused on single features such as adjectives, adverbs, and verbs, while other techniques examined types of these features, such as comparative adjectives, superlative adjectives, or general adverbs. Furthermore, POS as linguistic entities have also been studied and extensively classified by researchers, such as CLAWS-C7. For sentiment analysis, none of the studies conceptualized all possible POS features under similar conditions to draw firm conclusion. This research is centered on the following objectives: 1) examining the impact of various types of adjectives and adverbs that have not been previously explored for sentiment classification; 2) analyzing potential combinations of adjectives and adverbs types 3) conducting a comparison with a benchmark dataset for better classification accuracy. To assess the concept, a renowned human annotated dataset of tweets is investigated. Results showed that classification accuracy for adjectives is improved up to 83% based on the general superlative adjective whereas for adverbs, comparative general adverb also depicted significant accuracy improvement. Their combination with general adjectives and general adverbs also played a substantial role. The unexplored potential of adjectives and adverb types proved better in accuracy against state-of-the-art probabilistic model. In comparison to lexicon-based model, proposed research model overruled the dependency of lexicon-based dictionary where each term first needs to be matched for semantic orientation. The evident outcomes also help in time reduction aspect where huge volume of data need to be processed swiftly. This noteworthy contribution brought up significant knowledge and direction for domain experts. In the future, the proposed technique will be explored for other types of textual data across different domains.

'Laughing at us'

That Donald J. Trump won the US Presidential Election of 2016 defied the expectations of almost all seasoned observers of the US political system. Scholarly explanations stress structural factors that produced a substantial cohort of identity-vulnerable voters in tandem with Trump’s personal populist appeal. Trump benefitted from his political outsider status, celebrity familiarity, heteronormative masculinity, and unconventional rhetorical style including humour, all of which were amplified (and also mocked) by extensive mainstream news media coverage. Trump’s distinctive political use of the social media micro-blogging platform Twitter (now known as X) in engaging those cohorts has also been emphasized. The research presented here sits at the confluence of prior scholarly work on structural causes of identity vulnerability, on contestation involving humour, and on the enhanced political usage of Twitter. It examines the historical record of Trump’s tweets and re-tweets to see if and how the demarcation of his ‘humour brand’ affectively engaged his substantial Twitter followers and may thereby have contributed to his electoral success in 2016.

SENTIMENT ANALYSIS OF WEIBO COMMENTS BASED ON CONVOLUTIONAL NEURAL NETWORK

Sentiment analysis of social media comments plays a crucial role in understanding public opinion and sentiment trends. This study focuses on sentiment analysis of Weibo comments, a popular microblogging platform in China, using a (CNN) approach. We propose a novel methodology that leverages the inherent structural dependencies among comments and users in the Weibo network to capture nuanced sentiment patterns. By representing Weibo comments as a graph, with users and comments as nodes and their interactions as edges, we exploit the relational information to enhance sentiment classification accuracy. Furthermore, we employ attention mechanisms to prioritize influential users and comments in the sentiment analysis process. Through extensive experiments on real Weibo datasets, our proposed CNN-based sentiment analysis framework demonstrates superior performance compared to traditional methods, achieving high accuracy in sentiment classification tasks. This research contributes to advancing sentiment analysis techniques in social media platforms and provides valuable insights for understanding public sentiment dynamics in online communities like Weibo.

Cultural differences in microblogging: How Western IT Companies adapt Twitter (X) activities to the Chinese Weibo context

Companies worldwide use microblogs to communicate with stakeholders, but there may be cultural differences in how they do it. A content analysis was conducted comparing Twitter (currently known as X) and Weibo accounts of four Western IT companies; Weibo accounts of four similar Chinese companies served as benchmarks. Results show that Western microblog activities differed in many respects from Chinese practices. Specifically, they focused more on technology and less on marketing and community-building. In their localization strategies, Western companies chose to adapt (e.g., paying attention to community building), not adapt (e.g., keeping their technological profile), or cautiously adapt (e.g., using somewhat more emojis or reluctantly experimenting with sweepstakes). Cultural differences on microblogging platforms are comprehensive and multifaceted and cannot be easily reduced to established cultural dimensions. Processes of cultural adaptation, therefore, depend on profound knowledge of the business environment and cultural differences.

Odio-BERT: Evaluating domain task impact in hate speech detection

The rise of social media and micro-blogging platforms has led to concerns about hate speech, its potential to incite violence, psychological trauma, extremist beliefs, and self-harm. We have proposed a novel model, Odio-BERT for detecting hate speech using a pretrained BERT language model. This specialized model is specifically designed for detecting hate speech in the Spanish language, and when compared to existing models, it consistently outperforms them. The study provides valuable insights into addressing hate speech in the Spanish language and explores the impact of domain tasks.

Statement of Removal

This Statement of Removal refers to: 
 Mastodon Content Warnings: Inappropriate Contents in a Microblogging Platform

A STUDY ON SOCIAL MEDIA ADDICTION AMONG YOUTHS

Social media has become a widely used term today, referring to electronic communication platforms such as social networking websites and micro-blogging platforms. These platforms enable users to create online communities where they can share information, ideas, personal messages, videos, and more. While popular social media websites like Facebook, Twitter, Tumblr, Instagram, and WhatsApp have amassed millions of registered users, social media encompasses much more than just a sharing platform. It comprises interactive technologies that facilitate the creation and sharing of information, ideas, career interests, and various forms of expression within virtual communities and networks. Most individuals utilize social media networking sites to interact with both old and new friends, whether they are physical or online connections. The rapid evolution of technology has transformed the world into a global village, making communication more accessible than ever before. However, technology, like a coin, has both positive and negative impacts. Social media addiction, characterized by excessive concern and an uncontrollable urge to log on or use social media, can impair important aspects of life. This research paper aims to identify the types of social media platforms to which youths are addicted and analyze the positive and negative impacts on them. The data for this study will be collected from primary and secondary sources, and analytical presentations will be included.

Inferring cascading effects of natural and built environments failures from social media videos in disasters

The objective of this study is to infer the cascading effects between disruptive events in disasters using content coding and imagery representation techniques on social media videos. It is well recognized that disaster events are cascading due to the dynamic interactions between natural and built environments. Understanding why and how the failures of natural and built environments such as road and vehicle damages cascade is critical for retrospective investigation and disaster adaptation. Social media provides high-volume and real-time data, which has been broadly used in disaster management. What's more, the immense availability of social media videos, which record the facts of disaster events with high spatial and temporal resolutions, provides a unique opportunity to review and uncover the cascading process in disasters. Here, we propose an analytical method from two theoretical perspectives: content understanding and the philosophy of cascading effects. First, we establish a content classification system to code social media videos with respect to specific disaster events. Then, an imagery representation approach is developed to capture the cascading effects based on the observed relations from the social media videos. To examine the application of these methods and to derive new insights for the cascading effects, we leverage empirical social media video data collected from Weibo, a popular Chinese microblogging platform, during Typhoon Lekima. The coding results show that social media videos contain a wealth of information regarding disaster impacts, enabling us to unveil the occurrences and processes of the natural and built environment failures. By establishing cascading relations among disaster failures, we can present a whole picture of the complex disaster events induced by typhoons. The proposed method and findings of this study can contribute to not only computational modeling of cascading failures within natural and built environment, but also informing the design and planning of cities in adapting to future disasters.

War on feminism: an analysis of metaphorical representations on Weibo

ABSTRACT Social media has become a critical medium for disseminating key feminist ideas; however, negative representations of feminism have been observed on Chinese platforms. Metaphor analysis offers valuable insights into public perceptions and attitudes towards social issues, highlighting the role of metaphors in influencing reasoning and decision-making. This study delves into metaphors related to feminism on Weibo, a Chinese microblogging platform akin to Twitter. By analyzing 206,089 Weibo posts about feminism using established metaphor identification methods, six metaphorical categories were identified: War, business, illness, animal, religion, and other metaphors, with war metaphors being the most prevalent. Many of these metaphors carry negative implications, typically portraying feminism as irrational, exploitative, and disruptive to societal norms and values, leading to calls for its suppression. This study contributes to existing literature by highlighting the metaphorical framing of feminism on social media and offering novel insights into its evaluation within the Chinese context.

Towards low-resource rumor detection: Unified contrastive transfer with propagation structure

The truth is significantly hampered by massive rumors that spread along with breaking news or popular topics. Since there is sufficient corpus gathered from the same domain for model training, existing rumor detection algorithms show promising performance on yesterday’s news. However, due to a lack of substantial training data and prior expert knowledge, they are poor at spotting rumors concerning unforeseen events, especially those propagated in different languages (i.e., low-resource regimes). In this paper, we propose a simple yet effective framework with unified contrastive transfer learning, to detect rumors by adapting the features learned from well-resourced rumor data to that of the low-resourced with only few-shot annotations. More specifically, we first represent rumor circulated on social media as an undirected topology for enhancing the interaction of user opinions, and then train the propagation-structured model via a unified contrastive paradigm to mine effective clues simultaneously from both post semantics and propagation structure. Our model explicitly breaks the barriers of the domain and/or language issues, via language alignment and a novel domain-adaptive contrastive learning mechanism. To well-generalize the representation learning using a small set of annotated target events, we reveal that rumor-indicative signal is closely correlated with the uniformity of the distribution of these events. We design a target-wise contrastive training mechanism with three event-level data augmentation strategies, capable of unifying the representations by distinguishing target events. Extensive experiments conducted on four low-resource datasets collected from real-world microblog platforms demonstrate that our framework achieves much better performance than state-of-the-art methods and exhibits a superior capacity for detecting rumors at early stages.

End-to-End Deep Networks with Hierarchical Attention and Capsule Capabilities for Misinformation Detection on Microblogging Platforms

End-to-End Deep Networks with Hierarchical Attention and Capsule Capabilities for Misinformation Detection on Microblogging Platforms

Public Opinion About COVID-19 on a Microblog Platform in China: Topic Modeling and Multidimensional Sentiment Analysis of Social Media.

The COVID-19 pandemic raised wide concern from all walks of life globally. Social media platforms became an important channel for information dissemination and an effective medium for public sentiment transmission during the COVID-19 pandemic. Mining and analyzing social media text information can not only reflect the changes in public sentiment characteristics during the COVID-19 pandemic but also help the government understand the trends in public opinion and reasonably control public opinion. First, this study collected microblog comments related to the COVID-19 pandemic as a data set. Second, sentiment analysis was carried out based on the topic modeling method combining latent Dirichlet allocation (LDA) and Bidirectional Encoder Representations from Transformers (BERT). Finally, a machine learning linear regression (ML-LR) model combined with a sparse matrix was proposed to explore the evolutionary trend in public opinion on social media and verify the high accuracy of the model. The experimental results show that, in different stages, the characteristics of public emotion are different, and the overall trend is from negative to positive. The proposed method can effectively reflect the characteristics of the different times and space of public opinion. The results provide theoretical support and practical reference in response to public health and safety events.

A Study on the Application of Social Network Analysis Based Approach in the Optimization of Cultural Communication Paths and Efficiency

Abstract The optimization of cultural communication paths is an important part of Chinese culture’s continued dialogue with the world in the face of increasingly fierce international competition. In this paper, we use social network analysis to collect relevant data on the microblogging platform and analyze the overall network structure of “The Belt and Road Initiative” using network density, network distance, and clustering coefficients. Using social network analysis metrics, we calculate the node position of the network path of “The Belt and Road Initiative” and analyze the influence of each microblog account. We filter the cultural communication topics of “The Belt and Road Initiative”, scrutinize the participating regions, and establish the characteristics of the cooperation mode within this cultural communication. “The Belt and Road Initiative” has a network density of 0.247, 0.281, and 0.295, respectively, which indicates that the network density of these three topics is small and there is less information exchange, but the cohesive force of cultural communication is stronger. “People’s Daily,” “Central Committee of the Communist Youth League,” and “Phoenix Video” have high intermediary centrality (712.256, 700.425, and 695.253, respectively). The high intermediary centrality indicates that these nodes have a great influence on the flow of resources in the information dissemination network and can accelerate the efficiency of the “Belt and Road Initiative” cultural dissemination path.