Natural Language Processing for Depression Prediction on Sina Weibo: Method Study and Analysis.

Background: Depression represents a pressing global public health concern, impacting the physical and mental well-being of hundreds of millions worldwide. Notwithstanding advances in clinical practice, an alarming number of individuals at risk for depression continue to face significant barriers to timely diagnosis and effective treatment, thereby exacerbating a burgeoning social health crisis. Objective: This study seeks to develop a novel online depression risk detection method using natural language processing technology to identify individuals at risk of depression on the Chinese social media platform Sina Weibo. Methods: First, we collected approximately 527,333 posts publicly shared over 1 year from 1600 individuals with depression and 1600 individuals without depression on the Sina Weibo platform. We then developed a hierarchical transformer network for learning user-level semantic representations, which consists of 3 primary components: a word-level encoder, a post-level encoder, and a semantic aggregation encoder. The word-level encoder learns semantic embeddings from individual posts, while the post-level encoder explores features in user post sequences. The semantic aggregation encoder aggregates post sequence semantics to generate a user-level semantic representation that can be classified as depressed or nondepressed. Next, a classifier is employed to predict the risk of depression. Finally, we conducted statistical and linguistic analyses of the post content from individuals with and without depression using the Chinese Linguistic Inquiry and Word Count. Results: We divided the original data set into training, validation, and test sets. The training set consisted of 1000 individuals with depression and 1000 individuals without depression. Similarly, each validation and test set comprised 600 users, with 300 individuals from both cohorts (depression and nondepression). Our method achieved an accuracy of 84.62%, precision of 84.43%, recall of 84.50%, and F1-score of 84.32% on the test set without employing sampling techniques. However, by applying our proposed retrieval-based sampling strategy, we observed significant improvements in performance: an accuracy of 95.46%, precision of 95.30%, recall of 95.70%, and F1-score of 95.43%. These outstanding results clearly demonstrate the effectiveness and superiority of our proposed depression risk detection model and retrieval-based sampling technique. This breakthrough provides new insights for large-scale depression detection through social media. Through language behavior analysis, we discovered that individuals with depression are more likely to use negation words (the value of "swear" is 0.001253). This may indicate the presence of negative emotions, rejection, doubt, disagreement, or aversion in individuals with depression. Additionally, our analysis revealed that individuals with depression tend to use negative emotional vocabulary in their expressions ("NegEmo": 0.022306; "Anx": 0.003829; "Anger": 0.004327; "Sad": 0.005740), which may reflect their internal negative emotions and psychological state. This frequent use of negative vocabulary could be a way for individuals with depression to express negative feelings toward life, themselves, or their surrounding environment. Conclusions: The research results indicate the feasibility and effectiveness of using deep learning methods to detect the risk of depression. These findings provide insights into the potential for large-scale, automated, and noninvasive prediction of depression among online social media users.

Analyzing the impact of unemployment on mental health among Chinese university graduates: a study of emotional and linguistic patterns on Weibo.

Purpose: This study explores the intricate relationship between unemployment rates and emotional responses among Chinese university graduates, analyzing how these factors correlate with specific linguistic features on the popular social media platform Sina Weibo. The goal is to uncover patterns that elucidate the psychological and emotional dimensions of unemployment challenges among this demographic. Methods: The analysis utilized a dataset of 30,540 Sina Weibo posts containing specific keywords related to unemployment and anxiety, collected from January 2019 to June 2023. The posts were pre-processed to eliminate noise and refine the data quality. Linear regression and textual analyses were employed to identify correlations between unemployment rates for individuals aged 16-24 and the linguistic characteristics of the posts. Results: The study found significant fluctuations in urban youth unemployment rates, peaking at 21.3% in June 2023. A corresponding increase in anxiety-related expressions was noted in the social media posts, with peak expressions aligning with high unemployment rates. Linguistic analysis revealed that the category of "Affect" showed a strong positive correlation with unemployment rates, indicating increased emotional expression alongside rising unemployment. Other categories such as "Negative emotion" and "Sadness" also showed significant correlations, highlighting a robust relationship between economic challenges and emotional distress. Conclusion: The findings underscore the profound impact of unemployment on the emotional well-being of university students, suggesting that economic hardships are closely linked to psychological stress and heightened negative emotions. This study contributes to a holistic understanding of the socio-economic challenges faced by young adults, advocating for comprehensive support systems that address both the economic and psychological facets of unemployment.

Discursive delegitimisation of homosexuality on Chinese social media.

This article contributes to the understanding of public representations of homosexuality in China by focusing on the case of a homophobic textbook. College student Xixi sued Jinan University Press (JUP) in 2017 for classifying homosexuality as a psychosexual disorder. Three years later, a Chinese court dismissed Xixi's lawsuit against the allegedly homophobic textbook published by JUP. The ruling elicited responses on Chinese social media that demonstrated the polarisation of public opinion regarding homosexuality. This article investigates discursive representations of homosexuality in online space by analysing the public discourse surrounding this problem. Using van Leeuwen's discursive delegitimation strategies (i.e., authorisation, moral evaluation, rationalisation, and mythopoesis), 496 comments posted on Sina Weibo were employed and subjected to discourse analysis. According to our findings, these strategies contribute to public opposition to homosexuality, portrayed as unhealthy, infertile, disruptive, and corrosive. The article concludes by discussing the emerging sociocultural factors on Sina Weibo that influence the anti-homosexuality attitudes of Weibo users.

Exploring the motivations and obstacles of the public's garbage classification participation: evidence from Sina Weibo.

China has been implementing garbage classification to improve resource recycling for many years. Since garbage classification is essentially a social activity, it needs the active participation of the public. However, the phenomenon of "high practice, low effect" is widespread in most cities. Therefore, this paper uses the data from Sina Weibo to analyze the reasons for the poor garbage classification effect. First, the key factors affecting residents' willingness to participate in garbage classification are identified based on the text-mining method. Further, this paper analyzes the reasons that promote or hinder the residents' intention of garbage classification. Finally, the resident's attitude towards garbage classification is explored by the score of the text's emotional orientation, and further the reasons for the positive and negative emotional orientation are analyzed, respectively. The main conclusions are as follows: (1) The proportion of residents holding negative sentiment towards garbage classification is as high as 55%. (2) Residents' positive emotions are mainly caused by the public's sense of environmental protection inspired by publicity and education, and the incentive measures taken by the government. (3) The main reasons for negative emotions are imperfect infrastructure and unreasonable garbage sorting arrangements.

How health organizations communicate about COVID-19 on social media: a comparative content analysis.

BACKGROUND: This study examined how different health organizations (i.e., the Chinese CDC, the Korean CDC, the United States CDC, and WHO) communicated about the COVID-19 pandemic on social media, thus providing implications for organizations touse social media effectively in global health crises in the future. METHODS: Three bilingual researchers conducted a content analysis ofsocial media posts (N = 1,343) of these health organizations on Twitter and Sina Weibo to explore the frames of the COVID-19 pandemic, the purposes, and the strategies to communicate about it. RESULTS: Prevention was the dominant frame of the social media content of these four health organizations. Information update was the major communication purpose for WHO, the United States CDC, and the Korean CDC; however, guidance was the primary communication purpose for the Chinese CDC. The United States CDC, the Chinese CDC, and the Korean CDC heavily relied on multiple social media strategies (i.e., visual, hyperlink, and authority quotation) in their communication to the public about the COVID-19 pandemic, whereas WHO primarily employed quoting authorities. Significantdifferences were revealed across these health organizations in frames, communication purposes, and strategies. Theoretical and practical implications and limitations were discussed. CONCLUSIONS: This study examined how different global health organizations communicate about the COVID-19 pandemic on social media. We discussed how and why these global health organizations communicate the COVID-19 pandemic, which would help health-related organizations design messages strategically on global public health issues in the future.

Attention and Sentiment of the Chinese Public toward a 3D Greening System Based on Sina Weibo.

The rapid development of global urbanization over the years has led to a significant increase in the urban population, resulting in an imbalance in the urban green space structure. Transforming the urban 2D space green quantity into a 3D space green quantity to create 3D greenery systems (TGS) is a space resource that cannot be ignored in the process of urban green space expansion. This research gathered and analyzed Sina Weibo post information and user information related to TGS to investigate the changing trend of attention status and emotional orientation of the Chinese public on TGS. We employed web crawler technology and text mining to search and analyze the data on the Sina Weibo platform. This research aids policymakers and stakeholders in comprehending the general public's perspective on TGS and showing the transmission channel of public sentiment and the origins of negative sentiment. Results indicate that the public's attention to TGS has greatly increased since the shift in the government's idea of governance, although it still needs improvement. Despite TGS's good thermal insulation and air purification effects, 27.80% of the Chinese public has a negative attitude toward it. The public's negative sentiment of TGS housing is not solely due to pricing. The public is mainly concerned about the damage to the structure of buildings caused by TGS, the subsequent maintenance of plants, the increase in indoor mosquitoes, and lighting and humidity problems. This research helps decision makers understand the public opinion communication process via social media and provides corresponding solutions, which is of great significance for the future development of TGS.

Public Concern about Air Pollution and Related Health Outcomes on Social Media in China: An Analysis of Data from Sina Weibo (Chinese Twitter) and Air Monitoring Stations.

To understand the temporal variation, spatial distribution and factors influencing the public's sensitivity to air pollution in China, this study collected air pollution data from 2210 air pollution monitoring sites from around China and used keyword-based filtering to identify individual messages related to air pollution and health on Sina Weibo during 2017-2021. By analyzing correlations between concentrations of air pollutants (PM2.5, PM10, CO, NO2, O3 and SO2) and related microblogs (air-pollution-related and health-related), it was found that the public is most sensitive to changes in PM2.5 concentration from the perspectives of both China as a whole and individual provinces. Correlations between air pollution and related microblogs were also stronger when and where air quality was worse, and they were also affected by socioeconomic factors such as population, economic conditions and education. Based on the results of these correlation analyses, scientists can survey public concern about air pollution and related health outcomes on social media in real time across the country and the government can formulate air quality management measures that are aligned to public sensitivities.

Perspectives on Surrogacy in Chinese Social Media: A Content Analysis of Microblogs on Weibo.

Background: This article explores the social media discourse on transnational surrogacy and the issue of surrogacy more broadly considering recent news about the Chinese celebrity Zheng Shuang, which revealed that she had hired a surrogate mother in the United States and had later abandoned the surrogate babies. It aims to provide insight on how Chinese citizenry uses social media to express opinions on ethical and legal issues concerning surrogacy. Methods: We conducted a content analysis of microblogs from the social media platform Weibo posted within a month after the event was reported on January 17, 2021. The entire data set included 37,895 posts, which were analyzed for topic exploration using word frequency and keyword co-occurrence techniques, and a smaller sample of 1,000 posts was selected for an in-depth content analysis. Results: We established that the words "Zheng Shuang," "surrogacy," "babies," "abandoning babies," and "Zhang Heng" were most frequently used, with "law," "ethics," "justification," "legality," and "illegal" sharing high connections with these keywords. The qualitative content analysis further established that 399 microblogs (39.9%) expressed value judgements towards Zheng Shuang's surrogacy, and 61.9% (n=247) opposed her surrogacy, while only 7.0% (n=28) were supportive. The major reason (n=72) against the celebrity's surrogacy was that it was unfair and risky to surrogate children in this case. One hundred twenty-eight posts made value judgements towards surrogacy in principle, with 115 opposing surrogacy, and only two supportive posts. We also established that users with legal background had very limited presence in surrogacy discussions on Weibo, while users from healthcare professions did not engage at all in the social media debate. Conclusion: Opposition to surrogacy in Chinese social media discourse is primarily based on ethical and moral objections. The protection of surrogate children and surrogate women's rights was the major concerns expressed by social media users, suggesting that this issue would likely be at the center of a future public debate regarding the regulation of surrogacy. We found the lack of healthcare professionals' perspectives in social media discussions on Zheng's Surrogacy disconcerting and suggest their inclusion in public deliberations to ensure that the public is better educated, and substantive concerns are properly addressed.

Deep Hierarchical Ensemble Model for Suicide Detection on Imbalanced Social Media Data.

As a serious worldwide problem, suicide often causes huge and irreversible losses to families and society. Therefore, it is necessary to detect and help individuals with suicidal ideation in time. In recent years, the prosperous development of social media has provided new perspectives on suicide detection, but related research still faces some difficulties, such as data imbalance and expression implicitness. In this paper, we propose a Deep Hierarchical Ensemble model for Suicide Detection (DHE-SD) based on a hierarchical ensemble strategy, and construct a dataset based on Sina Weibo, which contains more than 550 thousand posts from 4521 users. To verify the effectiveness of the model, we also conduct experiments on a public Weibo dataset containing 7329 users' posts. The proposed model achieves the best performance on both the constructed dataset and the public dataset. In addition, in order to make the model applicable to a wider population, we use the proposed sentence-level mask mechanism to delete user posts with strong suicidal ideation. Experiments show that the proposed model can still effectively identify social media users with suicidal ideation even when the performance of the baseline models decrease significantly.

Examining Crisis Communication Using Semantic Network and Sentiment Analysis: A Case Study on NetEase Games.

The mobile game "Immortal Conquest," created by NetEase Games, caused a dramatic user dissatisfaction event after an introduction of a sudden and uninvited "pay-to-win" update. As a result, many players filed grievances against NetEase in a court. The official game website issued three apologies, with mix results, to mitigate the crisis. The goal of the present study is to understand user feedback content from the perspective of Situational Crisis Communication Theory through semantic network analysis and sentiment analysis to explore how an enterprise's crisis communication strategy affects users' attitudes. First, our results demonstrate that the diminishing crisis communication strategies (excuse and justification) do not change players' negative attitudes. It was not a failure because it successfully alleviated the players' legal complaints and refocused their attention on the game itself. Second, the rebuild (apology & compensation) strategy was effective because it significantly increased the percentage of positive emotions and regenerated expectations for the game. The litigation crisis was identified within gamer communications with respect to Chinese gaming companies for the first time. Nevertheless, this does not indicate an increase in overall legal awareness among the larger Chinese population. It may only reflect greater legal awareness among Chinese online gamers. Fourth, gamers emphasized that they and enterprises should be equally involved when communicating with each other. Finally, in-game paid items should be reasonably priced, otherwise, they will drive users to competitors.

China public emotion analysis under normalization of COVID-19 epidemic: Using Sina Weibo.

The prevention and control of the coronavirus disease 2019 (COVID-19) epidemic in China has entered a phase of normalization. The basis for evaluating and improving public health strategies is understanding the emotions and concerns of the public. This study establishes a fine-grained emotion-classification model to annotate the emotions of 32,698 Sina Weibo posts related to COVID-19 prevention and control from July 2022 to August 2022. The Dalian University of Technology (DLUT) emotion-classification system was adjusted to form four pairs (eight categories) of bidirectional emotions: good-disgust, joy-sadness, anger-fear, and surprise-anticipation. A lexicon-based method was proposed to classify the emotions of Weibo posts. Based on the selected Weibo posts, the present study analyzed the Chinese public's sentiments and emotions. The results showed that positive sentiment accounted for 51%, negative sentiment accounted for 24%, and neutral sentiment accounted for 25%. Positive sentiments were dominated by good and joy emotions, and negative sentiments were dominated by fear and disgust emotions. The proportion of positive sentiments on official Weibo (accounts belonging to government departments and official media) is significantly higher than that on personal Weibo. Official Weibo users displayed a weak guiding effect on personal users in terms of positive sentiment and the two groups of users were almost completely synchronized in terms of negative sentiment. The linear discriminant analysis (LDA) was performed on the two negative emotions of fear and disgust in the personal posts. The present study found that the emotion of fear was mainly related to COVID-19 infection and death, control of people with positive nucleic acid tests, and the outbreak of local epidemic, while the emotion of disgust was mainly related to the long-term existence of the epidemic, the cost of nucleic acid tests, non-implementation of prevention and control measures, and the occurrence of foreign epidemics. These findings suggest that Chinese attitudes toward epidemic prevention and control are positive and optimistic; however, there is also a notable proportion of fear and disgust. It is expected that this study will help public health administrators to evaluate the effectiveness of possible countermeasures and work toward precise prevention and control of the COVID-19 epidemic.

Exploring the impact of Chuangguandong Movement on individualism in China based on Sina Weibo information.

The voluntary frontier settlement hypothesis holds that frontier movements can promote the formation of individualism in the frontier area. The Chuangguandong Movement is one of China's voluntary frontier movements that potentially had a positive impact on the formation of high individualism in the northeastern provinces. Previous studies used independent/interdependent measures of self-construal scale, symbolic self-inflation, nepotism tasks, and percentage of most common names, to examine the differences in the independence between Heilongjiang and Shandong residents, which may be related to the Chuangguandong Movement. However, these studies were limited by certain factors such as sample size and objectivity of materials acquisition. In this study, we obtained Sina Weibo big data for period 2010-2020 to overcome the limitation of previous work. Using text feature extraction and keyword word frequency calculation methods based on the individualism/collectivism dictionary, we found that the level of individualism in Northeast China was higher than that in Shandong Province, which was consistent with previous research. Through the discussion of the four representative theoretical frameworks of individualism, the voluntary frontier settlement theory was considered as a potential explanation for the high degree of individualism in Northeast China.

China Public Psychology Analysis About COVID-19 Under Considering Sina Weibo Data.

COVID-19 not only poses a huge threat to public health, but also affects people's mental health. Take scientific and effective psychological crisis intervention to prevent large-scale negative emotional contagion is an important task for epidemic prevention and control. This paper established a sentiment classification model to make sentiment annotation (positive and negative) about the 105,536 epidemic comments in 86 days on the official Weibo of People's Daily, the test results showed that the accuracy of the model reached 88%, and the AUC value was greater than 0.9. Based on the marked data set, we explored the potential law between the changes in Internet public opinion and epidemic situation in China. First of all, we found that most of the Weibo users showed positive emotions, and the negative emotions were mainly caused by the fear and concern about the epidemic itself and the doubts about the work of the government. Secondly, there is a strong correlation between the changes of epidemic situation and people's emotion. Also, we divided the epidemic into three period. The proportion of people's negative emotions showed a similar trend with the number of newly confirmed cases in the growth and decay period, and the extinction period. In addition, we also found that women have more positive emotional performance than men, and the high-impact groups is also more positive than the low-impact groups. We hope that these conclusions can help China and other countries experiencing severe epidemics to guide publics respond.

What Happens When People with Depression Gather Online?

Depression is a common mental disease that impacts people of all ages and backgrounds. To meet needs that cannot otherwise be met, people with depression or who tend to suffer from depression often gather in online depression communities. However, since joining a depression community exposes members to the depression of others, the impact of such communities is not entirely clear. This study therefore explored what happens when people with depression gather in Sina Weibo's Depression Super Topic online community. Through website crawling, postings from Depression Super Topic were compared with postings from members' regular timelines with respect to themes, emotions disclosed, activity patterns, and the number of likes and comments. Topics of distilled postings covering support, regulations, emotions and life sharing, and initiating discussions were then coded. From comparison analysis, it was found that postings in the Depression Super Topic community received more comments and disclosed more emotions than regular timelines and that members were more active in the community at night. This study offers a picture of what occurs when people with depression gather online, which helps better understand their issues and therefore provide more targeted support.

Longitudinal Change of Mental Health among Active Social Media Users in China during the COVID-19 Outbreak.

During the COVID-19 pandemic, every day, updated case numbers and the lasting time of the pandemic became major concerns of people. We collected the online data (28 January to 7 March 2020 during the COVID-19 outbreak) of 16,453 social media users living in mainland China. Computerized machine learning models were developed to estimate their daily scores of the nine dimensions of the Symptom Checklist-90 (SCL-90). Repeated measures analysis of variance (ANOVA) was used to compare the SCL-90 dimension scores between Wuhan and non-Wuhan residents. Fixed effect models were used to analyze the relation of the estimated SCL-90 scores with the daily reported cumulative case numbers and lasting time of the epidemic among Wuhan and non-Wuhan users. In non-Wuhan users, the estimated scores for all the SCL-90 dimensions significantly increased with the lasting time of the epidemic and the accumulation of cases, except for the interpersonal sensitivity dimension. In Wuhan users, although the estimated scores for all nine SCL-90 dimensions significantly increased with the cumulative case numbers, the magnitude of the changes was generally smaller than that in non-Wuhan users. The mental health of Chinese Weibo users was affected by the daily updated information on case numbers and the lasting time of the COVID-19 outbreak.

Emotional Attitudes of Chinese Citizens on Social Distancing During the COVID-19 Outbreak: Analysis of Social Media Data.

BACKGROUND: Wuhan, China, the epicenter of the COVID-19 pandemic, imposed citywide lockdown measures on January 23, 2020. Neighboring cities in Hubei Province followed suit with the government enforcing social distancing measures to restrict the spread of the disease throughout the province. Few studies have examined the emotional attitudes of citizens as expressed on social media toward the imposed social distancing measures and the factors that affected their emotions. OBJECTIVE: The aim of this study was twofold. First, we aimed to detect the emotional attitudes of different groups of users on Sina Weibo toward the social distancing measures imposed by the People's Government of Hubei Province. Second, the influencing factors of their emotions, as well as the impact of the imposed measures on users' emotions, was studied. METHODS: Sina Weibo, one of China's largest social media platforms, was chosen as the primary data source. The time span of selected data was from January 21, 2020, to March 24, 2020, while analysis was completed in late June 2020. Bi-directional long short-term memory (Bi-LSTM) was used to analyze users' emotions, while logistic regression analysis was employed to explore the influence of explanatory variables on users' emotions, such as age and spatial location. Further, the moderating effects of social distancing measures on the relationship between user characteristics and users' emotions were assessed by observing the interaction effects between the measures and explanatory variables. RESULTS: Based on the 63,169 comments obtained, we identified six topics of discussion-(1) delaying the resumption of work and school, (2) travel restrictions, (3) traffic restrictions, (4) extending the Lunar New Year holiday, (5) closing public spaces, and (6) community containment. There was no multicollinearity in the data during statistical analysis; the Hosmer-Lemeshow goodness-of-fit was 0.24 (χ28=10.34, P>.24). The main emotions shown by citizens were negative, including anger and fear. Users located in Hubei Province showed the highest amount of negative emotions in Mainland China. There are statistically significant differences in the distribution of emotional polarity between social distancing measures (χ220=19,084.73, P<.001), as well as emotional polarity between genders (χ24=1784.59, P<.001) and emotional polarity between spatial locations (χ24=1659.67, P<.001). Compared with other types of social distancing measures, the measures of delaying the resumption of work and school or travel restrictions mainly had a positive moderating effect on public emotion, while traffic restrictions or community containment had a negative moderating effect on public emotion. CONCLUSIONS: Findings provide a reference point for the adoption of epidemic prevention and control measures, and are considered helpful for government agencies to take timely actions to alleviate negative emotions during public health emergencies.

COVID-19 Sentiment and the Chinese Stock Market: Evidence from the Official News Media and Sina Weibo.

This study quantitatively measures the Chinese stock market's reaction to sentiments regarding the Novel Coronavirus 2019 (COVID-19). Using 6.3 million items of textual data extracted from the official news media and Sina Weibo blogsite, we develop two COVID-19 sentiment indices that capture the moods related to COVID-19. Our sentiment indices are real-time and forward-looking indices in the stock market. We discover that stock returns and turnover rates were positively predicted by the COVID-19 sentiments during the period from December 17, 2019 to March 13, 2020. Consistent with this prediction, margin trading and short selling activities intensified proactively with growth sentiment. Overall, these results illustrate how the effects of the pandemic crisis were amplified by the sentiments.

Network Structure and Community Evolution Online: Behavioral and Emotional Changes in Response to COVID-19.

Background: The measurement and identification of changes in the social structure in response to an exceptional event like COVID-19 can facilitate a more informed public response to the pandemic and provide fundamental insights on how collective social processes respond to extreme events. Objective: In this study, we built a generalized framework for applying social media data to understand public behavioral and emotional changes in response to COVID-19. Methods: Utilizing a complete dataset of Sina Weibo posts published by users in Wuhan from December 2019 to March 2020, we constructed a time-varying social network of 3.5 million users. In combination with community detection, text analysis, and sentiment analysis, we comprehensively analyzed the evolution of the social network structure, as well as the behavioral and emotional changes across four main stages of Wuhan's experience with the epidemic. Results: The empirical results indicate that almost all network indicators related to the network's size and the frequency of social interactions increased during the outbreak. The number of unique recipients, average degree, and transitivity increased by 24, 23, and 19% during the severe stage than before the outbreak, respectively. Additionally, the similarity of topics discussed on Weibo increased during the local peak of the epidemic. Most people began discussing the epidemic instead of the more varied cultural topics that dominated early conversations. The number of communities focused on COVID-19 increased by nearly 40 percent of the total number of communities. Finally, we find a statistically significant "rebound effect" by exploring the emotional content of the users' posts through paired sample t-test (P = 0.003). Conclusions: Following the evolution of the network and community structure can explain how collective social processes changed during the pandemic. These results can provide data-driven insights into the development of public attention during extreme events.

A Novel Machine Learning Framework for Comparison of Viral COVID-19-Related Sina Weibo and Twitter Posts: Workflow Development and Content Analysis.

BACKGROUND: Social media plays a critical role in health communications, especially during global health emergencies such as the current COVID-19 pandemic. However, there is a lack of a universal analytical framework to extract, quantify, and compare content features in public discourse of emerging health issues on different social media platforms across a broad sociocultural spectrum. OBJECTIVE: We aimed to develop a novel and universal content feature extraction and analytical framework and contrast how content features differ with sociocultural background in discussions of the emerging COVID-19 global health crisis on major social media platforms. METHODS: We sampled the 1000 most shared viral Twitter and Sina Weibo posts regarding COVID-19, developed a comprehensive coding scheme to identify 77 potential features across six major categories (eg, clinical and epidemiological, countermeasures, politics and policy, responses), quantified feature values (0 or 1, indicating whether or not the content feature is mentioned in the post) in each viral post across social media platforms, and performed subsequent comparative analyses. Machine learning dimension reduction and clustering analysis were then applied to harness the power of social media data and provide more unbiased characterization of web-based health communications. RESULTS: There were substantially different distributions, prevalence, and associations of content features in public discourse about the COVID-19 pandemic on the two social media platforms. Weibo users were more likely to focus on the disease itself and health aspects, while Twitter users engaged more about policy, politics, and other societal issues. CONCLUSIONS: We extracted a rich set of content features from social media data to accurately characterize public discourse related to COVID-19 in different sociocultural backgrounds. In addition, this universal framework can be adopted to analyze social media discussions of other emerging health issues beyond the COVID-19 pandemic.

Detecting changes in attitudes toward depression on Chinese social media: A text analysis.

BACKGROUND & AIMS: Depression is a common and sometimes severe form of mental illness, and public attitudes towards depression can impact the psychological and social functioning of depressed patients. The purpose of the present study was to investigate public attitudes toward depression and three-year trends in these attitudes using big data analysis of social media posts in China. METHODS: A search of publically available Sina Weibo posts from January 2014 to July 2017 identified 20,129 hot posts with the keyword term "depression". We first used a Chinese Linguistic Psychological Text Analysis System (TextMind) to analyze linguistic features of the posts. And, then we used topic models to conduct semantic content analysis to identify specific themes in Weibo users' attitudes toward depression. RESULTS: Linguistic features analysis showed a significant increase over time in the frequency of terms related to affect, positive emotion, anger, cognition (including the subcategory of insight), and conjunctions. Semantic content analysis identified five common themes: severe effects of depression, stigma, combating stigma, appeals for understanding, and providing support. There was a significant increase over time in references to social (as opposed to professional) support, and a significant decrease over time in references to the severe consequences of depression. CONCLUSIONS: Big data analysis of Weibo posts is likely to provide less biased information than other methods about the public's attitudes toward depression. The results suggest that although there is ongoing stigma about depression, there is also an upward trend in mentions of social support for depressed persons. A supervised learning statistical model can be developed in future research to provide an even more precise analysis of specific attitudes.