Using text mining and sentiment analysis to analyse YouTube Italian videos concerning vaccination

The circular economy through the prism of machine learning and the YouTube video media platform

The transition to a Circular Economy (CE) is rapidly gaining ground across countries and industries. It is the means of achieving more sustainable development by adopting innovative environmentally friendly strategies and saving primary resources. There are several studies indicating the increasing public and corporate interest in the CE but still remain limited in terms of the multitude and utilization of social media data. This work aims to shed light on the most common topics discussed on the YouTube platform, related to the CE. For this reason, we selected 17 videos including the term "Circular Economy" since these have been the most relevant with a sufficient number of comments and views. The model identified two main topics referring to "Sustainable industry and environmental responsibility" and "Circular Economy and resource management" which is a strong indicator of the people's interest in the utilization of resources alongside industrial and corporate activities. The two-topic configuration presented the highest coherence score; however, five and ten-topic configurations have been deployed since there was no extreme differentiation in the model's performance, which could provide more detailed insights. This work's innovation lies in utilizing Machine Learning techniques and social media data to unravel CE's debates.

Understanding the determinants of vaccine hesitancy in the United States: A comparison of social surveys and social media

The COVID-19 pandemic prompted governments worldwide to implement a range of containment measures, including mass gathering restrictions, social distancing, and school closures. Despite these efforts, vaccines continue to be the safest and most effective means of combating such viruses. Yet, vaccine hesitancy persists, posing a significant public health concern, particularly with the emergence of new COVID-19 variants. To effectively address this issue, timely data is crucial for understanding the various factors contributing to vaccine hesitancy. While previous research has largely relied on traditional surveys for this information, recent sources of data, such as social media, have gained attention. However, the potential of social media data as a reliable proxy for information on population hesitancy, especially when compared with survey data, remains underexplored. This paper aims to bridge this gap. Our approach uses social, demographic, and economic data to predict vaccine hesitancy levels in the ten most populous US metropolitan areas. We employ machine learning algorithms to compare a set of baseline models that contain only these variables with models that incorporate survey data and social media data separately. Our results show that XGBoost algorithm consistently outperforms Random Forest and Linear Regression, with marginal differences between Random Forest and XGBoost. This was especially the case with models that incorporate survey or social media data, thus highlighting the promise of the latter data as a complementary information source. Results also reveal variations in influential variables across the five hesitancy classes, such as age, ethnicity, occupation, and political inclination. Further, the application of models to different MSAs yields mixed results, emphasizing the uniqueness of communities and the need for complementary data approaches. In summary, this study underscores social media data's potential for understanding vaccine hesitancy, emphasizes the importance of tailoring interventions to specific communities, and suggests the value of combining different data sources.

Flu vaccination coverage in Italy in the COVID-19 era: A fuzzy functional k-means (FFKM) approach

BACKGROUND

In Europe, flu vaccination coverage has decreased, and there are complex barriers to overcome to vaccinate against flu. Many studies have been conducted to estimate vaccination coverage. The COVID-19 pandemic threatens to disrupt immunization programs in many countries, including Italy, where vaccination against the flu is recommended but not mandatory. This paper aims to understand changes in flu vaccine uptake in Italian regions.

METHODS

Using functional data analysis and fuzzy functional k-means clustering, we investigated changes in flu vaccine coverage in Italian regions before (2010-2019) and after (2020-2022) the COVID-19 vaccination period.

RESULTS

The period of COVID-19 pandemic brought an increase in vaccine coverage globally. Elbow's method determined that the optimal number of clusters in vaccination uptake is 2. Apulia, Basilicata, Emilia Romagna, Liguria, Molise, Tuscany, and Umbria in 2019 belong less to the group with low flu vaccination uptake (G1) but increase their tendency to belong to this group over time: they decrease their propensity to be vaccinated for flu. For others, it seems that COVID-19 served as a push to increase flu vaccination coverage rates. Sicily appears to be the region that has responded best to the pandemic, changing its membership value from 2019 to 2022.

CONCLUSION

The present study highlights that the COVID-19 era has resulted in a higher flu vaccination coverage rate. Moreover, the regional level's improvement or worsening in flu vaccination coverage rate is not affected by the historical gap and socio-cultural and economic differences prevailing among Italian regions.

Spreading of misinformation on mass media and digital platforms regarding vaccines. A systematic scoping review on stakeholders, policymakers, and sentiments/behavior of Italian consumers

Studies on traditional and social media have found that misinformation about vaccines has been widely spread over the last decade, negatively impacting public opinion and people's willingness to get vaccinated. We reviewed the sentiments of Italian users to define the characteristic of anti-vax and pro-vax contents and defined the strategies to deal with the misinformation. Scopus, MEDLINE/PubMed, Google Scholar (up to page 10), and ISI Web of Knowledge databases were systematically searched. Research articles, brief reports, commentaries, and letters published between January 1, 2010 and March 30, 2022 were included in the search. No-vax or ambiguous contents in Italian mass media are not prevalent compared to neutral and pro-vax content; the communication of no-vax groups is significantly simplified, favoring the understanding of the topics by users. Events related to vaccinations are associated with news coverage by media, search engine consultations, and user reactions on social networks. In this context, the activity of no-vax groups is triggered, and misinformation and fake news spread even further. A multifactorial approach is necessary to manage online user sentiment and use mass and social media as health promotion tools.

Coordinated inauthentic behavior: An innovative manipulation tactic to amplify COVID-19 anti-vaccine communication outreach via social media

Coordinated inauthentic behavior (CIB) is a manipulative communication tactic that uses a mix of authentic, fake, and duplicated social media accounts to operate as an adversarial network (AN) across multiple social media platforms. The article aims to clarify how CIB's emerging communication tactic "secretly" exploits technology to massively harass, harm, or mislead the online debate around crucial issues for society, like the COVID-19 vaccination. CIB's manipulative operations could be one of the greatest threats to freedom of expression and democracy in our society. CIB campaigns mislead others by acting with pre-arranged exceptional similarity and "secret" operations. Previous theoretical frameworks failed to evaluate the role of CIB on vaccination attitudes and behavior. In light of recent international and interdisciplinary CIB research, this study critically analyzes the case of a COVID-19 anti-vaccine adversarial network removed from Meta at the end of 2021 for brigading. A violent and harmful attempt to tactically manipulate the COVID-19 vaccine debate in Italy, France, and Germany. The following focal issues are discussed: (1) CIB manipulative operations, (2) their extensions, and (3) challenges in CIB's identification. The article shows that CIB acts in three domains: (i) structuring inauthentic online communities, (ii) exploiting social media technology, and (iii) deceiving algorithms to extend communication outreach to unaware social media users, a matter of concern for the general audience of CIB-illiterates. Upcoming threats, open issues, and future research directions are discussed.

Characterizing discourses about COVID-19 vaccines on Twitter: a topic modeling and sentiment analysis approach

BACKGROUND

Evidence-based health communication is crucial for facilitating vaccine-related knowledge and addressing vaccine hesitancy. To that end, it is important to understand the discourses about COVID-19 vaccination and attend to the publics' emotions underlying those discourses.

METHODS

We collect tweets related to COVID-19 vaccines from March 2020 to March 2021. In total, 304,292 tweets from 134,015 users are collected. We conduct a Latent Dirichlet Allocation (LDA) modeling analysis and a sentiment analysis to analyze the discourse themes and sentiments.

RESULTS

This study identifies seven themes of COVID-19 vaccine-related discourses. Vaccine advocacy (24.82%) is the most widely discussed topic about COVID-19 vaccines, followed by vaccine hesitancy (22.29%), vaccine rollout (12.99%), vaccine facts (12.61%), recognition for healthcare workers (12.47%), vaccine side effects (10.07%), and vaccine policies (4.75%). Trust is the most salient emotion associated with COVID-19 vaccine discourses, followed by anticipation, fear, joy, sadness, anger, surprise, and disgust. Among the seven topics, vaccine advocacy tweets are most likely to receive likes and comments, and vaccine fact tweets are most likely to receive retweets.

CONCLUSIONS

When talking about vaccines, publics' emotions are dominated by trust and anticipation, yet mixed with fear and sadness. Although tweets about vaccine hesitancy are prevalent on Twitter, those messages receive fewer likes and comments than vaccine advocacy messages. Over time, tweets about vaccine advocacy and vaccine facts become more dominant whereas tweets about vaccine hesitancy become less dominant among COVID-19 vaccine discourses, suggesting that publics become more confident about COVID-19 vaccines as they obtain more information.

On the development of an information system for monitoring user opinion and its role for the public

Social media services and analytics platforms are rapidly growing. A large number of various events happen mostly every day, and the role of social media monitoring tools is also increasing. Social networks are widely used for managing and promoting brands and different services. Thus, most popular social analytics platforms aim for business purposes while monitoring various social, economic, and political problems remains underrepresented and not covered by thorough research. Moreover, most of them focus on resource-rich languages such as the English language, whereas texts and comments in other low-resource languages, such as the Russian and Kazakh languages in social media, are not represented well enough. So, this work is devoted to developing and applying the information system called the OMSystem for analyzing users' opinions on news portals, blogs, and social networks in Kazakhstan. The system uses sentiment dictionaries of the Russian and Kazakh languages and machine learning algorithms to determine the sentiment of social media texts. The whole structure and functionalities of the system are also presented. The experimental part is devoted to building machine learning models for sentiment analysis on the Russian and Kazakh datasets. Then the performance of the models is evaluated with accuracy, precision, recall, and F1-score metrics. The models with the highest scores are selected for implementation in the OMSystem. Then the OMSystem's social analytics module is used to thoroughly analyze the healthcare, political and social aspects of the most relevant topics connected with the vaccination against the coronavirus disease. The analysis allowed us to discover the public social mood in the cities of Almaty and Nur-Sultan and other large regional cities of Kazakhstan. The system's study included two extensive periods: 10-01-2021 to 30-05-2021 and 01-07-2021 to 12-08-2021. In the obtained results, people's moods and attitudes to the Government's policies and actions were studied by such social network indicators as the level of topic discussion activity in society, the level of interest in the topic in society, and the mood level of society. These indicators calculated by the OMSystem allowed careful identification of alarming factors of the public (negative attitude to the government regulations, vaccination policies, trust in vaccination, etc.) and assessment of the social mood.

Effectiveness of Social Video Platforms in Promoting COVID-19 Vaccination Among Youth: A Content-Specific Analysis of COVID-19 Vaccination Topic Videos on Bilibili

Background

With the widespread promotion of the COVID-19 vaccination in China, videos about the vaccination have become increasingly available on social video platforms. With the User Generated Content model, different creators’ interpretations of COVID-19 vaccines may influence the attitudes towards the vaccines and vaccination.

Objective

To explore the overview of COVID-19 vaccine-related videos on Bilibili, discussing the communication effects of COVID-19 topic videos and its influencing factors.

Methods

A content analysis was applied to the 202 video samples obtained through data mining regarding the creator’s information, video presentation, and COVID-19 vaccine-related content.

Results

Individuals and medical professionals preferred VLOG videos, media chose to upload informational videos, and enterprises preferred to post showcase videos. Individuals were more likely to discuss the adverse reactions in their videos, while medical professionals were more likely to discuss the vaccination process for the COVID-19 vaccine. Videos with core issues positively influenced the video’s dissemination breadth. The attitudes toward the COVID-19 vaccine in the videos positively influence the recognition of the videos. The richness of knowledge points related to the COVID-19 vaccine negatively affected the recognition and participation.

Conclusion

Social video platforms could play an active role in the vaccination promotion for the youth. Health promotion-related departments and individuals could strengthen agenda setting, grasp the characteristics of young groups, and express positive attitudes toward health issues to achieve better health (vaccine) promotion.

Media Data and Vaccine Hesitancy: Scoping Review

Background

Media studies are important for vaccine hesitancy research, as they analyze how the media shapes risk perceptions and vaccine uptake. Despite the growth in studies in this field owing to advances in computing and language processing and an expanding social media landscape, no study has consolidated the methodological approaches used to study vaccine hesitancy. Synthesizing this information can better structure and set a precedent for this growing subfield of digital epidemiology.

Objective

This review aimed to identify and illustrate the media platforms and methods used to study vaccine hesitancy and how they build or contribute to the study of the media's influence on vaccine hesitancy and public health.

Methods

This study followed the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) guidelines. A search was conducted on PubMed and Scopus for any studies that used media data (social media or traditional media), had an outcome related to vaccine sentiment (opinion, uptake, hesitancy, acceptance, or stance), were written in English, and were published after 2010. Studies were screened by only 1 reviewer and extracted for media platform, analysis method, the theoretical models used, and outcomes.

Results

In total, 125 studies were included, of which 71 (56.8%) used traditional research methods and 54 (43.2%) used computational methods. Of the traditional methods, most used content analysis (43/71, 61%) and sentiment analysis (21/71, 30%) to analyze the texts. The most common platforms were newspapers, print media, and web-based news. The computational methods mostly used sentiment analysis (31/54, 57%), topic modeling (18/54, 33%), and network analysis (17/54, 31%). Fewer studies used projections (2/54, 4%) and feature extraction (1/54, 2%). The most common platforms were Twitter and Facebook. Theoretically, most studies were weak. The following five major categories of studies arose: antivaccination themes centered on the distrust of institutions, civil liberties, misinformation, conspiracy theories, and vaccine-specific concerns; provaccination themes centered on ensuring vaccine safety using scientific literature; framing being important and health professionals and personal stories having the largest impact on shaping vaccine opinion; the coverage of vaccination-related data mostly identifying negative vaccine content and revealing deeply fractured vaccine communities and echo chambers; and the public reacting to and focusing on certain signals-in particular cases, deaths, and scandals-which suggests a more volatile period for the spread of information.

Conclusions

The heterogeneity in the use of media to study vaccines can be better consolidated through theoretical grounding. Areas of suggested research include understanding how trust in institutions is associated with vaccine uptake, how misinformation and information signaling influence vaccine uptake, and the evaluation of government communications on vaccine rollouts and vaccine-related events. The review ends with a statement that media data analyses, though groundbreaking in approach, should supplement-not supplant-current practices in public health research.

Virtual Reality Technology: Analysis based on text and opinion mining

The purpose of this research is to highlight the importance of periodically analyzing the data obtained from the technological sources used by customers, such as user comments on social networks and videos, using qualitative data analysis software. This research analyzes user sentiments, words, and opinions about virtual reality (VR) videos on YouTube in order to explore user reactions to such videos, as well as to establish whether this technology contributes to the sustainability of natural environments. User-generated data can provide important information for decision making about future policies of companies that produce video content. The results of our analysis of 12 videos revealed that users predominantly perceived these videos positively. This conclusion was supported by the findings of an opinion and text analysis, which identified positive reviews for videos and channels with many followers and large numbers of visits. The features such as the quality of the video and the accessibility of technology were appreciated by the viewers, whereas videos that are 100% VR and require special glasses to view them do not have as many visits. However, VR was seen to be a product which viewers were interested in and, according to Google, there are an increasing number of searches and sales of VR glasses in holiday seasons. Emotions of wonder and joy are more evident than emotions of anger or frustration, so positive feelings can be seen to be predominant.

Technical report: an online international weight control registry to inform precision approaches to healthy weight management

BACKGROUND

Personalizing approaches to prevention and treatment of obesity will be a crucial aspect of precision health initiatives. However, in considering individual susceptibility to obesity, much remains to be learned about how to support healthy weight management in different population subgroups, environments and geographical locations.

SUBJECTS/METHODS

The International Weight Control Registry (IWCR) has been launched to facilitate a deeper and broader understanding of the spectrum of factors contributing to success and challenges in weight loss and weight loss maintenance in individuals and across population groups. The IWCR registry aims to recruit, enroll and follow a diverse cohort of adults with varying rates of success in weight management. Data collection methods include questionnaires of demographic variables, weight history, and behavioral, cultural, economic, psychological, and environmental domains. A subset of participants will provide objective measures of physical activity, weight, and body composition along with detailed reports of dietary intake. Lastly, participants will be able to provide qualitative information in an unstructured format on additional topics they feel are relevant, and environmental data will be obtained from public sources based on participant zip code.

CONCLUSIONS

The IWCR will be a resource for researchers to inform improvements in interventions for weight loss and weight loss maintenance in different countries, and to examine environmental and policy-level factors that affect weight management in different population groups. This large scale, multi-level approach aims to inform efforts to reduce the prevalence of obesity worldwide and its associated comorbidities and economic impacts.

TRIAL REGISTRATION

NCT04907396 (clinicaltrials.gov) sponsor SB Roberts; Tufts University IRB #13075.

Aspect Based Twitter Sentiment Analysis on Vaccination and Vaccine Types in COVID-19 Pandemic with Deep Learning

Due to the COVID-19 pandemic, vaccine development and community vaccination studies are carried out all over the world. At this stage, the opposition to the vaccine seen in the society or the lack of trust in the developed vaccine is an important factor hampering vaccination activities. In this study, aspect-base sentiment analysis was conducted for USA, UK, Canada, Turkey, France, Germany, Spain and Italy showing the approach of twitter users to vaccination and vaccine types during the COVID-19 period. Within the scope of this study, two datasets in English and Turkish were prepared with 928,402 different vaccine-focused tweets collected by country. In the classification of tweets, 4 different aspects (policy, health, media and other) and 4 different BERT models (mBERT-base, BioBERT, ClinicalBERT nad BERTurk) were used. 6 different COVID-19 vaccines with the highest frequency among the datasets were selected and sentiment analysis was made by using Twitter posts regarding these vaccines. To the best of our knowledge, this paper is the first attempt to understand people's views about vaccination and types of vaccines. With the experiments conducted, the results of the views of the people on vaccination and vaccine types were presented according to the countries. The success of the method proposed in this study in the F1 Score was between 84% and 88% in datasets divided by country, while the total accuracy value was 87%.

Changes of the Public Attitudes of China to Domestic COVID-19 Vaccination After the Vaccines Were Approved: A Semantic Network and Sentiment Analysis Based on Sina Weibo Texts

Introduction: On December 31, 2020, the Chinese government announced that the domestic coronavirus disease-2019 (COVID-19) vaccines have obtained approval for conditional marketing and are free for vaccination. This release drove the attention of the public and intense debates on social media, which reflected public attitudes to the domestic vaccine. This study examines whether the public concerns and public attitudes to domestic COVID-19 vaccines changed after the official announcement.

Methods: This article used big data analytics in the research, including semantic network and sentiment analysis. The purpose of the semantic network is to obtain the public concerns about domestic vaccines. Sentiment analysis reflects the sentiments of the public to the domestic vaccines and the emotional changes by comparing the specific sentiments shown on the posts before and after the official announcement.

Results: There exists a correlation between the public concerns about domestic vaccines before and after the official announcement. According to the semantic network analysis, the public concerns about vaccines have changed after the official announcement. The public focused on the performance issues of the vaccines before the official approval, but they cared more about the practical issues of vaccination after that. The sentiment analysis showed that both positive and negative emotions increased among the public after the official announcement. “Good” was the most increased positive emotion and indicated great public appreciation for the production capacity and free vaccination. “Fear” was the significantly increased negative emotion and reflected the public concern about the safety of the vaccines.

Conclusion: The official announcement of the approval for marketing improved the Chinese public acceptance of the domestic COVID-19 vaccines. In addition, safety and effectiveness are vital factors influencing public vaccine hesitancy.

Multi-perspectives systematic review on the applications of sentiment analysis for vaccine hesitancy

A substantial impediment to widespread Coronavirus disease (COVID-19) vaccination is vaccine hesitancy. Many researchers across scientific disciplines have presented countless studies in favor of COVID-19 vaccination, but misinformation on social media could hinder vaccination efforts and increase vaccine hesitancy. Nevertheless, studying people's perceptions on social media to understand their sentiment presents a powerful medium for researchers to identify the causes of vaccine hesitancy and therefore develop appropriate public health messages and interventions. To the best of the authors' knowledge, previous studies have presented vaccine hesitancy in specific cases or within one scientific discipline (i.e., social, medical, and technological). No previous study has presented findings via sentiment analysis for multiple scientific disciplines as follows: (1) social, (2) medical, public health, and (3) technology sciences. Therefore, this research aimed to review and analyze articles related to different vaccine hesitancy cases in the last 11 years and understand the application of sentiment analysis on the most important literature findings. Articles were systematically searched in Web of Science, Scopus, PubMed, IEEEXplore, ScienceDirect, and Ovid from January 1, 2010, to July 2021. A total of 30 articles were selected on the basis of inclusion and exclusion criteria. These articles were formed into a taxonomy of literature, along with challenges, motivations, and recommendations for social, medical, and public health and technology sciences. Significant patterns were identified, and opportunities were promoted towards the understanding of this phenomenon.

Characterizing HIV discussions and engagement on Twitter

The novel settings provided by social media facilitate users to seek and share information on a wide array of subjects, including healthcare and wellness. Analyzing health-related opinions and discussions on these platforms complement traditional public health surveillance systems to support timely and effective interventions. This study aims to characterize the HIV-related conversations on Twitter by identifying the prevalent topics and the key events and actors involved in these discussions. Through Twitter API, we collected tweets containing the hashtag #HIV for a one-year period. After pre-processing the collected data, we conducted engagement analysis, temporal analysis, and topic modeling algorithm on the analytical sample (n = 122,807). Tweets by HIV/AIDS/LGBTQ activists and physicians received the highest level of engagement. An upsurge in tweet volume and engagement was observed during global and local events such as World Aids Day and HIV/AIDS awareness and testing days for trans-genders, blacks, women, and the aged population. Eight topics were identified that include “stigma”, “prevention”, “epidemic in the developing countries”, “World Aids Day”, “treatment”, “events”, “PrEP”, and “testing”. Social media discussions offer a nuanced understanding of public opinions, beliefs, and sentiments about numerous health-related issues. The current study reports various dimensions of HIV-related posts on Twitter. Based on the findings, public health agencies and pertinent entities need to proactively use Twitter and other social media by engaging the public through involving influencers. The undertaken methodological choices may be applied to further assess HIV discourse on other popular social media platforms.

Prospective associations of regional social media messages with attitudes and actual vaccination: A big data and survey study of the influenza vaccine in the United States

•

Regional Twitter vaccine content was prospectively associated with attitudes.

•

Regional Twitter vaccine content was prospectively associated with vaccination.

•

Discussing the influenza vaccine with others can remove the negative effects of Twitter vaccine content.

Objective

Using longitudinal methods to assess regional associations between social media posts about vaccines and attitudes and actual vaccination against influenza in the US.

Methods

Geolocated tweets from U.S. counties (N = 115,330) were analyzed using MALLET LDA (Latent Dirichlet allocation) topic modeling techniques to correlate with prospective individual survey data (N = 3005) about vaccine attitudes, actual vaccination, and real-life discussions about vaccines with family and friends during the 2018–2019 influenza season.

Results

Ten topics were common across U.S. counties during the 2018–2019 influenza season. In the overall analyses, two of these topics (i.e., Vaccine Science Matters and Big Pharma) were associated with attitudes and behaviors. The topic concerning vaccine science in November-February was positively correlated with attitudes in February-March, r = 0.09, BF₁₀ = 3. Moreover, among respondents who did not discuss the influenza vaccine with family and friends, the topic about vaccine fraud and children in November-February was negatively correlated with attitudes in February-March and with vaccination in February-March, and April-May (rs = −0.18 to −0.25, BF₁₀ = 4–146). However, this was absent when participants had discussions about the influenza vaccine with family and friends.

Discussion

Regional vaccine content correlated with prospective measures of vaccine attitudes and actual vaccination.

Conclusions

Social media have demonstrated strong associations with vaccination patterns. When the associations are negative, discussions with family and friends appear to eliminate them. Programs to promote vaccination should encourage real-life conversations about vaccines.