Figure 3 - available via license: Creative Commons Attribution 4.0 International
Content may be subject to copyright.
Distribution of different types of sentiments across different time periods used for analysis. The bigger the heart symbol, the stronger the corresponding emotion was during the given time period.
Source publication
Background:
The ongoing COVID-19 pandemic increased the general public's anxiety, depression, post-traumatic stress disorder (PTSD), and psychological stress in various degrees worldwide. Better tailored mental health services and interventions cannot be achieved until we understand the patterns of mental health issues after disasters, especially...
Context in source publication
Similar publications
With the development of social media platforms such as Weibo, they have provided a broad platform for the expression of public sentiments during the pandemic. This study aims to explore the emotional attitudes of Chinese netizens toward the COVID-19 opening-up policies and their related thematic characteristics. Using Python, 145,851 texts were col...
Citations
... Second, our comment data were limited to posts that garnered more than 500 comments, a consequence of Weibo's anticrawling mechanism and the constraints of our Weibo crawler code, coupled with the fact that many com-ments have reading permissions set, which could potentially compromise the representativeness of our sample. Furthermore, our text mining was not conducted at fine-grained spatiotemporal dimensions, and the patterns of user behavior and emotional expressions could vary across different periods (Yu et al., 2021). Lastly, we also gathered data on likes, comments, reposts, and user personal information, and future research will delve into a more detailed examination of information dissemination paths and influencing factors. ...
Global earthquake resilience enhancement necessitates a cohesive international community and proactive public engagement. The ubiquity of social media, and its significant user data generation, positions it as a formidable tool in emergency management. This article scrutinizes the 2023 Türkiye–Syria Earthquake, employing an analysis of over 150,000 Weibo posts to assess Chinese netizens’ attention toward this nondomestic disaster. Findings reveal substantial social media engagement, with a distinct disparity between posts and comments. Posts provide a comprehensive chronicle of the earthquake response, encompassing disaster reporting, Chinese rescue teams, international emergency response, and life‐saving efforts. Comments serve as a conduit for emotional expression, focusing on prayers for the disaster‐stricken area, the tragic loss of a Turkish TV producer, expressions of sympathy and donations, the sense of helplessness evoked by natural disasters, and prayers for the rescue team and response comparison. The public engagement evolved from initial symbolic participation to more substantive involvement in the later stages, reflecting the process of both public engagement and the enhancement of resilience awareness. Sentiment and emotion analysis indicate that, subsequent to the initial 3‐day period dedicated to disaster reporting, the sentiment within the posts was predominantly positive, while comments exhibited a more negative sentiment, primarily characterized by “sadness” and “disgust.” Spatial analysis revealed amplified attention toward the disaster in Sichuan Province and other regions with superior internet access. The article also discerned that comments exhibited a notably shorter lifespan than posts. This article proffers a novel perspective on the analysis of foreign disasters and furnishes invaluable insights for the enhancement of global earthquake resilience through public engagement.
... In the early stages of the pandemic, fear, anxiety, and anger became the main tone of public sentiment, which was closely related to the uncertainty of the epidemic's spread, the strain on medical resources, and the chaos of information dissemination (23). However, with the gradual implementation of epidemic prevention and control measures and the enhancement of information transparency, the public's sentiment has gradually shifted toward hope and optimism (24). This emotional shift not only reflects society's adaptability to the pandemic but also underscores the importance of government and social resource management (25). ...
With the development of social media platforms such as Weibo, they have provided a broad platform for the expression of public sentiments during the pandemic. This study aims to explore the emotional attitudes of Chinese netizens toward the COVID-19 opening-up policies and their related thematic characteristics. Using Python, 145,851 texts were collected from the Weibo platform. Sentiment analysis and topic modeling techniques were employed to reveal the distribution of public emotions and key themes. The study found that the proportions of emotions were as follows: Good (46%), Happy (11%), Anger (17%), Disgust (6%), Sadness (10%), Surprise (2%), and Fear (8%). Through topic analysis, the following main themes were identified: medical resource shortages, healthcare workers, national policies, and COVID-19 sequelae. Based on the results of sentiment and topic analysis, public emotions toward the COVID-19 opening-up policies were categorized into three dimensions: panic, trust, and acceptance. Panic was primarily associated with medical resource shortages, concerns about COVID-19 sequelae, and doubts about policy transparency and fairness. Trust was reflected in public gratitude toward healthcare workers and support for national policies. Acceptance represented the public’s optimism about returning to normal life. The findings demonstrate that changes in public emotions not only reflect the social impact of policy implementation but also highlight the critical roles of medical resource allocation, information transparency, and psychological health support in adjusting pandemic policies. This study provides empirical evidence and theoretical support for the government to develop more precise pandemic control strategies.
... Afterward, two master students of English majors were invited as independent raters to assess the website's outputs based on the numbers. Additionally, intensity of Emotion (IOE) was extracted using SnowNLP library in Python, which is one of the most popular Python libraries for sentiment analysis for Chinese language natural language processing (Yu et al., 2021). ...
ChatGPT has been demonstrated to possess significant capabilities in generating intricate human-like text, and recent studies have established that its performance in theory of mind (ToM) tasks is strikingly comparable to a nine-year-old child’s. However, it remains unknown whether ChatGPT outperforms children of this age group in Chinese writing, a task credibly related to ToM. To justify the claim, this study compared ChatGPT with nine-year-old children in making Chinese compositions (i.e., science-themed and nature-themed narratives), aiming to unveil the relative advantages and disadvantages by human writers and ChatGPT in Chinese writing. Based on the evaluative framework comprising of four indices (i.e., fluency, accuracy, complexity, and cohesion) to test writing quality, this study added an often-overlooked index “emotion” to extend the framework. Afterward, we collected 120 writing samples produced by ChatGPT and children and used the confirmatory factor analysis (CFA) and structural equation modelling (SEM) for data analysis and comparison. The results revealed that this age group of children surpassed ChatGPT in fluency and cohesion while ChatGPT transcended the children in accuracy. With respect to complexity, the children exhibited better skills in science-themed writing, but ChatGPT better in nature-themed writing. Most importantly, this study unlocked the pioneering discovery that children display more potent emotional expressions than ChatGPT in Chinese writing, providing an instance of evidence that ChatGPT is really even poorer than a nine-year-old child in ToM to some extent.
... During the pandemic, social media data became an essential tool for analyzing public sentiment (Heffner et al. 2021;Wang et al. 2022;Box-Steffensmeier and Moses, 2021;Xia et al. 2023), playing a vital role in understanding public reactions to containment policies (Tsao et al. 2021) and focusing on issues of public concern (Ashokkumar and Pennebaker, 2021;Abd-Alrazaq et al. 2020;De Santis et al. 2020). Analyzing the content of social media tweets and user interactions can provide valuable insights for public health policy, improve emergency response, and prepare for future infectious disease threats and other public health emergencies Yu et al. 2021). ...
This study examines the dynamic relationship between China’s COVID-19 containment policies and public sentiment, focusing on the significant lockdowns in Wuhan and Shanghai. We employed natural language processing (NLP) on Weibo text data to uncover how people’s emotions towards these containment measures changed over time and space. Our analysis reveals a critical evolution in public sentiment, transitioning from initial support to growing dissatisfaction, highlighting the impact of ‘pandemic fatigue’ and the socio-economic factors influencing these shifts. This study contributes to understanding the complex interplay between public health strategies and societal reactions, providing practical insights into the spatial variations of sentiment across different demographic and socio-economic groups. By elucidating the causal effects of containment policies on public sentiment and the subsequent rise in public skepticism, our research offers valuable lessons for policymakers in tailoring communication and interventions to mitigate negative public perceptions and foster compliance during health crises.
... This can be due to its role in processing and characterizing a wide range of opinions in a short period of time. In addition, the traditional method for extracting and examining public's emotions requires a large-scale survey, which is not always possible (Yu et al. 2021). Social media sites provide the means for users to express and share their opinions online. ...
The recognition of eye disorders has the potential to reduce blindness in people. The need for a procedural method is important to boost the overall recognition process. Although the identification of certain disease symptoms is crucial to an early diagnosis, this study proposed a procedural mechanism to predict eye diseases on the Twitter platform using users’ sentiments embedded in their social media data. Glaucoma was investigated as one example of various eye diseases. Themes related to glaucoma were extracted using Latent Dirichlet Allocation. Subsequently, association rules mining was employed to identify disease-related symptoms within each theme. Our results showed that certain emotions, such as fear and sadness emotions, were highly associated with glaucoma messages. The findings revealed that emotion-related features have a significant impact on improving the prediction process of glaucoma in patients. As a result, this study proposes a low-cost procedural mechanism for the early-stage detection of eye disorders using microblogs data. The proposed approach can advance current efforts toward developing clinical decision support systems capable of detecting diseases online.
... Various studies have proposed new sentiment analysis methods and compared existing tools (eg, TextBlob [74], VADER [12], and Stanza [13]) on topics related to COVID-19, mainly extracted from social media [6,16,17,[75][76][77][78]. However, to the best of our knowledge, there are no studies that have compared several sentiment analysis tools on health-related surveys-a more structured type of text data than social media posts-that collected knowledge, beliefs, and habits during the COVID-19 pandemic [79][80][81][82][83][84]. ...
Background
Health care providers and health-related researchers face significant challenges when applying sentiment analysis tools to health-related free-text survey data. Most state-of-the-art applications were developed in domains such as social media, and their performance in the health care context remains relatively unknown. Moreover, existing studies indicate that these tools often lack accuracy and produce inconsistent results.
Objective
This study aims to address the lack of comparative analysis on sentiment analysis tools applied to health-related free-text survey data in the context of COVID-19. The objective was to automatically predict sentence sentiment for 2 independent COVID-19 survey data sets from the National Institutes of Health and Stanford University.
Methods
Gold standard labels were created for a subset of each data set using a panel of human raters. We compared 8 state-of-the-art sentiment analysis tools on both data sets to evaluate variability and disagreement across tools. In addition, few-shot learning was explored by fine-tuning Open Pre-Trained Transformers (OPT; a large language model [LLM] with publicly available weights) using a small annotated subset and zero-shot learning using ChatGPT (an LLM without available weights).
Results
The comparison of sentiment analysis tools revealed high variability and disagreement across the evaluated tools when applied to health-related survey data. OPT and ChatGPT demonstrated superior performance, outperforming all other sentiment analysis tools. Moreover, ChatGPT outperformed OPT, exhibited higher accuracy by 6% and higher F-measure by 4% to 7%.
Conclusions
This study demonstrates the effectiveness of LLMs, particularly the few-shot learning and zero-shot learning approaches, in the sentiment analysis of health-related survey data. These results have implications for saving human labor and improving efficiency in sentiment analysis tasks, contributing to advancements in the field of automated sentiment analysis.
... However, in the research to date, less attention has been paid to mental health impacts and outcomes 88 and more research has been focused on the gauging and interpretation of public sentiment. [89][90][91] Research reveals differences in public BMJ Global Health focus and emotional states in different geographical regions. For example, SinaWeibo users in areas experiencing severe outbreaks have higher platform engagement levels and report poorer emotional states than users on other platforms, which is likely a reflection of the distress experienced by the community in acute outbreak phases. ...
Importance
The onset of the COVID-19 global pandemic highlighted the increasing role played by social media in the generation, dissemination and consumption of outbreak-related information.
Objective
The objective of the current review is to identify and summarise the role of social media in public health crises caused by infectious disease, using a five-step scoping review protocol.
Evidence review
Keyword lists for two categories were generated: social media and public health crisis. By combining these keywords, an advanced search of various relevant databases was performed to identify all articles of interest from 2000 to 2021, with an initial retrieval date of 13 December 2021. A total of six medical and health science, psychology, social science and communication databases were searched: PubMed, Web of Science, Scopus, Embase, PsycINFO and CNKI. A three-stage screening process against inclusion and exclusion criteria was conducted.
Findings
A total of 338 studies were identified for data extraction, with the earliest study published in 2010. Thematic analysis of the role of social media revealed three broad themes: surveillance monitoring, risk communication and disease control. Within these themes, 12 subthemes were also identified. Within surveillance monitoring, the subthemes were disease detection and prediction, public attitude and attention, public sentiment and mental health. Within risk communication, the subthemes were health advice, information-seeking behaviour, infodemics/misinformation circulation, seeking help online, online distance education and telehealth. Finally, within disease control, the subthemes were government response, public behaviour change and health education information quality. It was clear that the pace of research in this area has gradually increased over time as social media has evolved, with an explosion in attention following the outbreak of COVID-19.
Conclusions and relevance
Social media has become a hugely powerful force in public health and cannot be ignored or viewed as a minor consideration when developing public health policy. Limitations of the study are discussed, along with implications for government, health authorities and individual users. The pressing need for government and health authorities to formalise evidence-based strategies for communicating via social media is highlighted, as well as issues for individual users in assessing the quality and reliability of information consumed on social media platforms.
... Recent studies have demonstrated that the incidence of psychological problems increased in the general population during the COVID-19 pandemic [20][21][22][23][24][25]. Self-quarantine further aggravates the occurrence of psychological disorders [26][27][28][29][30][31][32][33]. ...
Background
This study aimed to evaluate the prevalence of anxiety and depressive symptoms among quarantined college students at school in Shanghai 2022 lockdown during the COVID-19 pandemic and investigate the association of gastrointestinal discomfort related-factors and skipping breakfast with anxiety and depressive symptoms.
Methods
384 quarantined college students in Shanghai China were recruited in this cross-sectional study from April 5th to May 29th, 2022. Generalized Anxiety Disorder (GAD-7) and Patient Health Questionnaire (PHQ-9) were used to assess anxiety and depressive symptoms, respectively.
Results
The prevalence of anxiety and depressive symptoms were 56.8% and 62.8%, respectively. Longer quarantine duration, higher education level, skipping breakfast, stomachache or abdominal pain, and nausea or dyspepsia were significantly associated with anxiety symptoms. Moreover, longer quarantine duration, being woman, skipping breakfast, stomachache or abdominal pain, and nausea or dyspepsia were markedly related to depressive symptoms. Notably, regularly physical exercising and taking positive attitude towards COVID-19 were negatively correlated with anxiety and depressive symptoms.
Conclusions
More attention should be paid to anxiety and depressive symptoms of quarantined college students and universities should provide timely psychological monitoring and intervention services to mitigate the impact of negative emotions on students. Effectively relieving gastrointestinal symptoms, insisting on eat breakfast, regularly exercising, and taking a positive attitude towards to COVID-19 might contribute to preventing the anxiety and depressive symptoms for those college students experiencing a long-term quarantine.
... Objective central policy: To ensure that the voices and aspirations of the public can be heard, considered, and implemented in making policies that impact their lives them (Chetty and Alathur 2020). Utilization of social media in policy makes it possible for inhabitants to participate in a manner active in various stages of the policy process, starting from planning and formulation until implementation and evaluation (Yu, Eisenman, and Han 2021). Through social media, individuals can disclose views they give input, deliver complaints, and discuss with fellow inhabitants or authorized parties. ...
This study aims to analyze the role of social media in using public policy on health services in the literature Scopus indexed. Deep social media government has become more critical in e-government. Considering social media for taking the policy, the government is the internal medium to push public policy on services and health to increase the government's performance. This research analyzed 454 Scopus database documents from 2018-2023 using "health service," "policy," and "service use" as keywords. The data was filtered using bibliometrics based on the relevance of keywords, author's country, and year of publication, limited to the last five years. Information is saved in RIS format and processed through device soft Citespace. CiteSpace software is used for publication data visualization and government plan formulation. Effective social media policy requires good management for public decision-making. The study analyzed only one topic on Scopus without using international index databasesI'll summarize three research by presenting papers based on criteria like publications, origin countries, fields of study, authors, institutions, issues, and citations. Medicine studies with percentages are dominant, King's College London leads, Draheim contributes the most, and the US is the biggest contributor.
... A sentiment analysis was conducted of the obtained review texts using the SnowNLP library in Python. SnowNLP is a tool library based on a Bayesian model speci cally designed for Chinese natural language processing and sentiment analysis [52]. The formula for the sentiment analysis was as follows: ...
Urban parks are essential components of urban ecosystems, providing vital ecological resources for city residents. However, the rapid expansion of high-density urban areas has led to an unequal distribution of park resources, raising growing concerns about spatial equity. To address these challenges, we employed an improved Gaussian two-step floating catchment area (2SFCA) method, considering park quality variations and integrating sentiment scores from park reviews to calculate a comprehensive park accessibility index, accounting for both supply and demand dynamics among park users. The results demonstrate the significance of park management, as users prioritise convenience and cleanliness of public facilities. Recreational quality significantly influences park distribution equity, with areas near Beijing’s initial greenbelt zone showing improved accessibility (IA). Nonetheless, our analysis exposes disparities in urban park resource allocation within the Chaoyang District, indicating relative inequity. Spatial supply and demand mismatches, especially in the northwest and southeast, are evident. To enhance park layout equity, we recommend strategies like identifying and repurposing underused spaces, establishing pocket parks and micro-green areas, and improving recreational facilities. It is crucial to address the needs of vulnerable groups such as older residents and children. These insights stress the importance of ensuring fair urban park access to enhance the well-being of all city residents.
