Public Perception of the COVID-19 Pandemic on Twitter: Sentiment Analysis and Topic Modeling Study
Open accessPublished version of a preprint
The DOI matches its Crossref record; 43 of the 43 references it cites were found in a registry or an index.
Abstract
Background: COVID-19 is a scientifically and medically novel disease that is not fully understood because it has yet to be consistently and deeply studied. Among the gaps in research on the COVID-19 outbreak, there is a lack of sufficient infoveillance data. Objective: The aim of this study was to increase understanding of public awareness of COVID-19 pandemic trends and uncover meaningful themes of concern posted by Twitter users in the English language during the pandemic. Methods: Data mining was conducted on Twitter to collect a total of 107,990 tweets related to COVID-19 between December 13 and March 9, 2020. The analyses included frequency of keywords, sentiment analysis, and topic modeling to identify and explore discussion topics over time. A natural language processing approach and the latent Dirichlet allocation algorithm were used to identify the most common tweet topics as well as to categorize clusters and identify themes based on the keyword analysis. Results: The results indicate three main aspects of public awareness and concern regarding the COVID-19 pandemic. First, the trend of the spread and symptoms of COVID-19 can be divided into three stages. Second, the results of the sentiment analysis showed that people have a negative outlook toward COVID-19. Third, based on topic modeling, the themes relating to COVID-19 and the outbreak were divided into three categories: the COVID-19 pandemic emergency, how to control COVID-19, and reports on COVID-19. Conclusions: Sentiment analysis and topic modeling can produce useful information about the trends in the discussion of the COVID-19 pandemic on social media as well as alternative perspectives to investigate the COVID-19 crisis, which has created considerable public awareness. This study shows that Twitter is a good communication channel for understanding both public concern and public awareness about COVID-19. These findings can help health departments communicate information to alleviate specific public concerns about the disease.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Figures











Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“First, most were based on tweets collected in the early stages of the pandemic, which were analyzed from an infoveillance perspective, thereby limiting their analyses to health-related responses (e.g., Abd-Alrazaq et al., 2020; Boon-Itt et al., 2020; Xue et al., 2020).”
From Understanding user responses to the COVID-19 pandemic on Twitter from a terror management theory perspective: Cultural differences among the US, UK and India (2022) · cites this paper as Boon-Itt et al., 2020
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“most research has been conducted from an infoveillance perspective, thus limiting our understanding to health-related responses, such as the origin of the novel coronavirus, preventive measures, and other public knowledge about the virus (Abd-Alrazaq et al., 2020; Boon-Itt et al., 2020; Chandrasekaran et al., 2020; Xue et al., 2020).”
From Understanding user responses to the COVID-19 pandemic on Twitter from a terror management theory perspective: Cultural differences among the US, UK and India (2022) · cites this paper as Boon-Itt et al., 2020
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
The epidemiology and pathogenesis of coronavirus disease (COVID-19) outbreak
Rothan, H. A., & Byrareddy, S. N. (2020) · Journal of Autoimmunity · doi:10.1016/j.jaut.2020.102433 · Cited by 5,840 · cited 1×
“These common symptoms, including fever and cough, remain consistent across several studies [35].”
ResultDetecting influenza epidemics using search engine query data
Ginsberg, J., Mohebbi, M. H., Patel, R. S., Brammer, L., Smolinski, M. S., & Brilliant, L. (2009) · Nature · doi:10.1038/nature07634 · Cited by 4,295
SARS and MERS: recent insights into emerging coronaviruses
de Wit, E., van Doremalen, N., Falzarano, D., & Munster, V. J. (2016) · Nature Reviews Microbiology · doi:10.1038/nrmicro.2016.81 · Cited by 3,510
Adhikari, S. P., Meng, S., Wu, Y.-J., Mao, Y.-P., Ye, R.-X., Wang, Q.-Z., Sun, C., Sylvia, S., Rozelle, S., Raat, H., et al. (2020) · Infectious Diseases of Poverty · doi:10.1186/s40249-020-00646-x · Cited by 2,402 · cited 1×
“On January 30, 2020, the WHO reported that COVID-19 was a public health emergency of international concern [4].”
BackgroundLi, S., Wang, Y., Xue, J., Zhao, N., & Zhu, T. (2020) · International Journal of Environmental Research and Public Health · doi:10.3390/ijerph17062032 · Cited by 1,835 · cited 1×
“The use of social media information to analyze syndromic surveillance, focusing on public health–related concerns using web-based information and content, is essential [10].”
BackgroundPandemics in the Age of Twitter: Content Analysis of Tweets during the 2009 H1N1 Outbreak
Chew, C., & Eysenbach, G. (2010) · PLoS ONE · doi:10.1371/journal.pone.0014118 · Cited by 1,614 · cited 1×
“Text analysis of Twitter data is one of the most important areas of focus in medical informatics research [8].”
BackgroundInfluenza Pandemics of the 20th Century
Kilbourne, E. D. (2006) · Emerging Infectious Diseases · doi:10.3201/eid1201.051254 · Cited by 1,224 · cited 1×
“In the course of history, there have been many infectious disease outbreaks in the human population, causing both loss of life and damage to economies [1].”
BackgroundCan we contain the COVID-19 outbreak with the same measures as for SARS?
Wilder-Smith, A., Chiew, C. J., & Lee, V. J. (2020) · The Lancet Infectious Diseases · doi:10.1016/s1473-3099(20)30129-8 · Cited by 954 · cited 1×
“Previous studies have also shown that prevention and control procedures, including quarantine, as well as reports on confirmed cases and medical treatments were major themes during previous disease outbreaks [42-45].”
ResultNussbaumer-Streit, B., Mayr, V., Dobrescu, A. I., Chapman, A., Persad, E., Klerings, I., Wagner, G., Siebert, U., Christof, C., Zachariah, C., et al. (2020) · Cochrane Database of Systematic Reviews · doi:10.1002/14651858.cd013574 · Cited by 937
Coronavirus Disease 19 (COVID-19): Implications for Clinical Dental Care
Ather, A., Patel, B., Ruparel, N. B., Diogenes, A., & Hargreaves, K. M. (2020) · Journal of Endodontics · doi:10.1016/j.joen.2020.03.008 · Cited by 901 · cited 1×
“COVID-19 is a new infectious disease that is spread by respiratory droplets and contact and is generally infectious to human beings [2].”
BackgroundClinical progression of patients with COVID-19 in Shanghai, China
Chen, J., Qi, T., Liu, L., Ling, Y., Qian, Z., Li, T., Li, F., Xu, Q., Zhang, Y., Xu, S., et al. (2020) · Journal of Infection · doi:10.1016/j.jinf.2020.03.004 · Cited by 789 · cited 1×
“[33,34].”
Top concerns of tweeters during the COVID-19 pandemic: Infoveillance study
Abd-Alrazaq, A., Alhuwail, D., Househ, M., Hamdi, M., & Shah, Z. (2020) · Journal of Medical Internet Research · doi:10.2196/19016 · Cited by 774 · cited 1×
“Abd-Alrazaq et al [15] conducted an infoveillance study on aspects of the COVID-19 pandemic, aiming to study the main topics of discussion related to the disease.”
BackgroundChen, E., Lerman, K., & Ferrara, E. (2020) · JMIR Public Health and Surveillance · doi:10.2196/19273 · Cited by 762 · cited 1×
“Chen et al [16] presented basic statistics that tracked only Twitter activity responding and reacting to COVID-19–related events.”
BackgroundExtracting, Computing and Exploring the Parameters of Statistical Models using R
Lüdecke, D., Ben-Shachar, M., Patil, I., & Makowski, D. (2020) · Journal of Open Source Software · doi:10.21105/joss.02445 · Cited by 518 · cited 1×
“In this study, we used the “tidy()” method, which is included in the broom package [28].”
MethodGeldsetzer, P. (2020) · Journal of Medical Internet Research · doi:10.2196/18790 · Cited by 504 · cited 1×
“The two key symptoms of COVID-19 are cough and fever [30]; common symptoms also include headaches and sneezing.”
BackgroundLi, L., Zhang, Q., Wang, X., Zhang, J., Wang, T., Gao, T.-L., Duan, W., Tsoi, K. K.-F., & Wang, F.-Y. (2020) · IEEE Transactions on Computational Social Systems · doi:10.1109/tcss.2020.2980007 · Cited by 483 · cited 1×
“Studies have also indicated that people pay attention to and search for disease-related words as the spread of infectious disease changes [41].”
BackgroundGlobal Sentiments Surrounding the COVID-19 Pandemic on Twitter: Analysis of Twitter Trends
Lwin, M. O., Lu, J., Sheldenkar, A., Schulz, P. J., Shin, W., Gupta, R., & Yang, Y. (2020) · JMIR Public Health and Surveillance · doi:10.2196/19447 · Cited by 443 · cited 1×
“Lwin et al [17] studied Twitter users’ public emotional responses to COVID-19, focusing on four basic emotions only.”
BackgroundShah, K., Kamrai, D., Mekala, H., Mann, B., Desai, K., & Patel, R. S. (2020) · Cureus · doi:10.7759/cureus.7405 · Cited by 425 · cited 1×
“Previous studies have also shown that prevention and control procedures, including quarantine, as well as reports on confirmed cases and medical treatments were major themes during previous disease outbreaks [42-45].”
ResultCOVID-19 Public Sentiment Insights and Machine Learning for Tweets Classification
Samuel, J., Ali, G. G. M. N., Rahman, M. M., Esawi, E., & Samuel, Y. (2020) · Information · doi:10.3390/info11060314 · Cited by 407
Chinese Public's Attention to the COVID-19 Epidemic on Social Media: Observational Descriptive Study
Zhao, Y., Cheng, S., Yu, X., & Xu, H. (2020) · Journal of Medical Internet Research · doi:10.2196/18825 · Cited by 364 · cited 1×
“This result is in accordance with a previous study [31], which explained the different stages of the public’s attention to COVID-19.”
ResultCharles-Smith, L. E., Reynolds, T. L., Cameron, M. A., Conway, M., Lau, E. H. Y., Olsen, J. M., Pavlin, J. A., Shigematsu, M., Streichert, L. C., Suda, K. J., et al. (2015) · PLOS ONE · doi:10.1371/journal.pone.0139701 · Cited by 324 · cited 1×
“Social media data text mining has been used to track diseases and assess public awareness concerning health issues, enabling disease forecasting [7].”
BackgroundLiu, Q., Zheng, Z., Zheng, J., Chen, Q., Liu, G., Chen, S., Chu, B., Zhu, H., Akinwunmi, B., Huang, J., et al. (2020) · Journal of Medical Internet Research · doi:10.2196/19118 · Cited by 288 · cited 1×
“Previous studies have also shown that prevention and control procedures, including quarantine, as well as reports on confirmed cases and medical treatments were major themes during previous disease outbreaks [42-45].”
ResultRosenbloom, L. (2019) · The Charleston Advisor · doi:10.5260/chara.21.2.8 · Cited by 284
Content Sharing in a Social Broadcasting Environment: Evidence from Twitter1
Shi, Z., Rui, H., & Whinston, A. B. (2014) · MIS Quarterly · doi:10.25300/misq/2014/38.1.06 · Cited by 283 · cited 1×
“Our purpose was to measure the intensity of Twitter activities relating to COVID-19 with the stipulated keywords based on tweets that received retweets, representing conversation and content sharing about COVID-19 on Twitter [23].”
BackgroundConversations and Medical News Frames on Twitter: Infodemiological Study on COVID-19 in South Korea
Park, H. W., Park, S., & Chong, M. (2020) · Journal of Medical Internet Research · doi:10.2196/18897 · Cited by 252 · cited 1×
“Moreover, Park et al [14] addressed information transmission networks and news-sharing behaviors on Twitter regarding COVID-19 in Korea only.”
BackgroundAnalysing the Sentiments towards Work-From-Home Experience during COVID-19 Pandemic
Dubey, A. D., & Tripathi, S. (2020) · Journal of Innovation Management · doi:10.24840/2183-0606_008.001_0003 · Cited by 174 · cited 1×
“The keywords related to public awareness were different according to the stage of the spread; this includes public emotion, which was mostly more negative than positive, where fear was the most negative word [36,37].”
BackgroundSesagiri Raamkumar, A., Tan, S. G., & Wee, H. L. (2020) · Journal of Medical Internet Research · doi:10.2196/19334 · Cited by 166
Shen, C., Chen, A., Luo, C., Zhang, J., Feng, B., & Liao, W. (2020) · Journal of Medical Internet Research · doi:10.2196/19421 · Cited by 157 · cited 1×
“Taking research published in 2020 as examples, Shen et al [12] studied mentions of symptoms and diseases on social media to predict COVID-19 case counts, and Huang et al [13] analyzed social media posts to study the characteristics of COVID-19 patients in China.”
MethodLwin, M. O., Lu, J., Sheldenkar, A., & Schulz, P. J. (2018) · International Journal of Environmental Research and Public Health · doi:10.3390/ijerph15091974 · Cited by 141 · cited 1×
“It is necessary to avoid undermining the possibility of a serious outbreak during the incubation stage [32].”
BackgroundUsing twitter and web news mining to predict COVID-19 outbreak
Rahmanian, V., & Jahanbin, K. (2020) · Asian Pacific Journal of Tropical Medicine · doi:10.4103/1995-7645.279651 · Cited by 141 · cited 1×
“Text mining can be used to extract health information from social media platforms such as Twitter [6].”
BackgroundHow people react to Zika virus outbreaks on Twitter? A computational content analysis
Fu, K., Liang, H., Saroha, N., Tse, Z. T. H., Ip, P., & Fung, I. C. (2016) · American Journal of Infection Control · doi:10.1016/j.ajic.2016.04.253 · Cited by 133 · cited 1×
“For example, during previous events such as natural disasters or disease outbreaks, people used Twitter to express their feelings [20,21]; this has been particularly true during the global epidemic of COVID-19.”
BackgroundSelf-reported COVID-19 symptoms on Twitter: an analysis and a research resource
Sarker, A., Lakamana, S., Hogg-Bremer, W., Xie, A., Al-Garadi, M. A., & Yang, Y. (2020) · Journal of the American Medical Informatics Association · doi:10.1093/jamia/ocaa116 · Cited by 133 · cited 1×
“Sentiment analysis using the National Research Council (NRC) sentiment lexicon enabled us to examine the expression of 10 terms related to basic emotions: anger , anticipation , disgust , fear , joy , negative , positive , sadness , surprise , and trust [26,27].”
MethodZika Virus–Related News Coverage and Online Behavior, United States, Guatemala, and Brazil
Southwell, B. G., Dolina, S. E., Jimenez-Magdaleno, K., Squiers, L., & Kelly, B. (2016) · Emerging infectious diseases · doi:10.3201/eid2207.160415 · Cited by 113 · cited 1×
“The result is in accordance with previous research showing that on social media, people’s interests are related to the latest news and major events regarding infectious diseases [40].”
ResultGuo, X., & Li, J. (2019) · International Conference on Social Networks Analysis, Management and Security (SNAMS) · doi:10.1109/snams.2019.8931720 · Cited by 97 · cited 1×
“Second, sentiment analysis, a natural language processing (NLP) approach, was used to categorize the sentiments appearing in Twitter messages [25].”
MethodUsage and analysis of Twitter during 2015 Chennai flood towards disaster management
Nair, M. R., Ramya, G. R., & Sivakumar, P. B. (2017) · Procedia Computer Science · doi:10.1016/j.procs.2017.09.089 · Cited by 93 · cited 1×
“For example, during previous events such as natural disasters or disease outbreaks, people used Twitter to express their feelings [20,21]; this has been particularly true during the global epidemic of COVID-19.”
BackgroundMining the Characteristics of COVID-19 Patients in China: Analysis of Social Media Posts
Huang, C., Xu, X., Cai, Y., Ge, Q., Zeng, G., Li, X., Zhang, W., Ji, C., & Yang, L. (2020) · Journal of Medical Internet Research · doi:10.2196/19087 · Cited by 89 · cited 1×
“[33,34].”
Crisis and Emergency Risk Communication
Heath, R. L. (2013) · Encyclopedia of Public Relations · doi:10.4135/9781452276236.n116 · Cited by 58
Mamidi, R., Miller, M., Banerjee, T., Romine, W. L., & Sheth, A. (2019) · JMIR Public Health and Surveillance · doi:10.2196/11036 · Cited by 36 · cited 1×
“In previous disease pandemics, negative sentiments were generally prevalent in social media [38].”
BackgroundPanda, M. (2018) · International Journal of Modern Education and Computer Science · doi:10.5815/ijmecs.2018.09.02 · Cited by 31 · cited 1×
“Stop words in English (eg, for , the , is ) were also removed, as were words such as corona or virus that may relate to other topics [24].”
BackgroundWerron, T., & Ringel, L. (2020) · Sociologica · doi:10.6092/issn.1971-8853/11172 · Cited by 29 · cited 1×
“COVID-19 is a scientifically and medically novel disease that is not fully understood, as it has yet to be consistently and deeply studied [9].”
BackgroundBartlett, C., & Wurtz, R. M. (2013) · Journal of Public Health Management and Practice · doi:10.1097/phh.0000000000000041 · Cited by 26 · cited 1×
“Twitter, one of these social media platforms, has often served as a communication modality during disease outbreaks [11].”
BackgroundGeographic variations in electronic cigarette advertisements on Twitter in the United States
Dai, H. D., Deem, M. J., & Hao, J. (2016) · International Journal of Public Health · doi:10.1007/s00038-016-0906-9 · Cited by 16 · cited 1×
“Data from Twitter are acceptable for research, being very rich and useful [19].”
(1900) · De Gruyter eBooks · doi:10.1515/9783111598697-003 · Cited by 1
Research works printed but not linked: 4
- Jahanbin, K, 2020, Asian Pac J of Trop Med 2020; 13, 13, 378
- Murray, C, ArXiv
- 10.2196/19087 · DOI
- Kleinberg, B, ArXiv
Cited by
What social media told us in the time of COVID-19: A scoping review
Tsao, S.-F., Chen, H., Tisseverasinghe, T., Yang, Y., Li, L., & Butt, Z. A. (2021) · The Lancet Digital Health · doi:10.1016/s2589-7500(20)30315-0 · Cited by 816
Survey on sentiment analysis: evolution of research methods and topics
Cui, J., Wang, Z., Ho, S.-B., & Cambria, E. (2023) · Artificial Intelligence Review · doi:10.1007/s10462-022-10386-z · Cited by 213
“Sentiment analysis on data from social media platforms related to COVID-19 has become a hot topic (Boon-Itt and Skunkan 2020).”
BackgroundQi, Y., & Shabrina, Z. (2023) · Social Network Analysis and Mining · doi:10.1007/s13278-023-01030-x · Cited by 193
“Sentiment analysis can be implemented to social media data to explore changes in people’s behavior, emotions, and opinions such as by dividing the spread trend of Covid-19 into three stages and exploring people’s negative sentiments toward Covid-19 based on topic modeling and feature extraction (Boon-Itt and Skunkan 2…”
Designing, Conducting, and Reporting Survey Studies: A Primer for Researchers
Zimba, O., & Gasparyan, A. Y. (2023) · Journal of Korean Medical Science · doi:10.3346/jkms.2023.38.e403 · Cited by 198
Cotfas, L., Delcea, C., Roxin, I., Ioanăş, C., Gherai, D. S., & Tajariol, F. (2021) · IEEE Access · doi:10.1109/access.2021.3059821 · Cited by 167
“The prevalence of negative sentiments over the positive ones in the case of the COVID-19 pandemic has been also underlined by Singh et al. [66], while Boon-Itt and Skunkan [67] have recorded a high discrepancy between the negative sentiments (covering 77.88% of tweets) and the positive sentiments (covering the rest of…”
BackgroundInfluential
Metadata
- Authors
- Boon-Itt, S., & Skunkan, Y.
- Affiliations
- Bangkok Christian Hospital (TH)
- Venue
- JMIR Public Health and Surveillance
- Publisher
- JMIR Publications Inc.
- Published
- 2020-11-11
- Volume
- 6
- Issue
- 4
- Pages
- e21978
- ISSN
- 2369-2960 (ISSN-L)
- PMC
- PMC7661106
- OpenAlex
- W3094221957
- Cited by
- 486
- References
- 43
- Citations by year
- 2026: 23 · 2025: 67 · 2024: 77 · 2023: 118 · 2022: 115 · 2021: 84 · 2020: 2
- FWCI
- 31.35
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- cc-by
- Locations
- JMIR Public Health and Surveillance (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Sentiment Analysis and Opinion Mining
- Social Sciences › Communication › Communication and COVID-19 Impact
- Health Sciences › Medicine › Epidemiology › Data-Driven Disease Surveillance
- Keywords
- COVID-19 pandemic, public perception, Twitter, social media discourse, public awareness, sentiment analysis, topic modeling, pandemic control measures, health communication, infoveillance
- MeSH terms
- COVID-19, Humans, Health Knowledge, Attitudes, Practice, Models, Psychological, Natural Language Processing, Data Mining, Pandemics, Social Media
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.