Social media platforms like X (formerly Twitter) generate vast volumes of user-generated content that provide real-time insights into public sentiment. Despite the widespread use of traditional machine learning methods, their limitations in capturing contextual nuances in noisy social media text remain a challenge. This study leverages the Sentiment140 dataset, comprising 1.6 million labeled tweets, and develops predictive models for binary sentiment classification using Naive Bayes, Logistic Regression, and the transformer-based BERT model. Experiments were conducted on a balanced subset of 12,000 tweets after comprehensive NLP preprocessing. Evaluation using accuracy, F1-score, and confusion matrices revealed that BERT significantly outperforms traditional models, achieving an accuracy of 89.5% and an F1-score of 0.89 by effectively modeling contextual and semantic nuances. In contrast, Naive Bayes and Logistic Regression demonstrated reasonable but consistently lower performance. To support practical deployment, we introduce SentiFeel, an interactive tool enabling real-time sentiment analysis. While resource constraints limited the dataset size and training epochs, future work will explore full corpus utilization and the inclusion of neutral sentiment classes. These findings underscore the potential of transformer models for enhanced public opinion monitoring, marketing analytics, and policy forecasting.

The national museum of a country, as a cultural symbol of the nation, plays an important role in cultural communication at home and abroad. This study explores the online brand identity construction of two national museums—the British Museum and the National Museum of China—to inform cultural brands of the discursive strategies to distinguish themselves from others and communicate with their audiences effectively. Informed by multimodal critical discourse analysis, this paper analyzes the websites of the two museums and their social media posts, depicts their brand identity prisms, and evaluates the effectiveness of their online communication. The results show that both museums use multimodal and hypertextual resources to create unique and congruent brand images in website design and social media interaction with their target audiences, fulfilling the institutional functions of museums as the symbol of national culture or world civilization. They express differential personalities and cultural values to reinforce their brand identities in different sociocultural and political contexts. The findings may provide insight into the use of multimodality in online communication for cultural institutions to enhance their brand images and promote cultural exchanges.

Despite the pervasive nature of internet use among adolescents and young adults, there is not enough knowledge about whether and how involvement in social media influences substance use patterns and the risk of drug use-related problems. This study was conducted to examine the complex relationship between substance use-related social media engagement (viewing, liking, commenting, and posting the substance use-related social media content) and the drug use-related problem in adolescents from public perception. We surveyed to determine the perception of social media users regarding the association between substance use-related social media content and substance/drug abuse problems. An anonymous online questionnaire was conducted to collect the response from each participant. The response was generated after collecting the data from 126 users of mixed ages. The data was stringently analyzed, and the response was displayed in the form of bar charts. The primary findings indicated a significant relationship between drug/alcohol-related social media engagement and drug/alcohol-related problems. From public perception, a positive correlation was found between the engagement in the drug use-related content and drug use associated problems. However, further research is needed to determine the right direction of these associations that can provide substantiative solutions for numerous interventions aiming to prevent drug use-related adverse consequences.

Background: Electronic cigarette (e-cigarette) use and vaping tobacco have increased rapidly worldwide, raising concerns about their health effects, social acceptability, and regulatory challenges. In many countries, e-cigarettes are more commonly used by individuals from higher socioeconomic status (SES) backgrounds, who, in theory, should have greater knowledge about e-cigarettes and their associated risks. However, it remains unclear why a group with more knowledge about e-cigarette risks would also hold more positive attitudes toward vaping and exhibit higher usage rates — a phenomenon that may represent a knowledge-behavior paradox. Understanding this paradox, along with the complex relationships between e-cigarette knowledge, attitudes, and behaviors, is critical for informing effective public health interventions, campaigns, social media messaging, and regulatory policies. Objectives: This study aimed to evaluate the complex relationship between SES, e-cigarette knowledge, pro-vaping attitudes, and e-cigarette use. Methods: The SMOKES Study (Study of Measurement of Knowledge and Examination of Support for Tobacco Control Policies) used a multi-center, cross-sectional design, collecting data from 2,403 college and university students across 15 provinces in Iran (covering nearly half of the country's provinces). The survey measured family income, age, sex, ethnicity, e-cigarette use, knowledge, and attitudes. Structural Equation Modeling (SEM) was employed to examine the interrelations between SES, knowledge, attitudes, and behavior, while adjusting for age, sex, and ethnic minority status. Results: SEM analysis confirmed the hypothesized paradox. Although greater knowledge about e-cigarettes was linked to less favorable attitudes toward vaping and lower use, pro-vaping attitudes emerged as the strongest predictor of vaping behavior, while knowledge played a weaker protective role. Notably, individuals with higher SES simultaneously showed higher knowledge and, paradoxically, more pro-e-cigarette attitudes and greater usage. Female students and ethnic minority students reported higher correct knowledge and lower pro-vaping attitudes and use. Although age and higher family income were associated with more favorable attitudes, they did not directly predict vaping behavior. These results suggest that for higher SES individuals, poor knowledge is not the main driver of e-cigarette use; rather, their pro-e-cigarette attitudes, which seem to outweigh the influence of knowledge, play a key role. Conclusions: Although individuals from higher SES backgrounds report greater correct knowledge about e-cigarettes, this knowledge does not necessarily translate into reduced positive attitudes or lower usage. This study highlights the complexity of these paradoxical effects and suggests that public health strategies need to go beyond simple education and knowledge-based interventions. Targeted approaches should address industry messaging, challenge misconceptions, and strengthen regulatory efforts to reduce e-cigarette use among young adults, including those from higher SES backgrounds.