Short CommunicationPublic sentiment on the global outbreak of monkeypox: an unsupervised machine learning analysis of 352,182 twitter posts
Introduction
Monkeypox is endemic to West and Central Africa, and before the current outbreaks, almost all cases outside of Africa were linked to international travel or through imported animals (zoonosis).1 In contrast, the current outbreak seems largely confined to homosexual and bisexual men, and the virus may have become more infectious or be capable of other modes or asymptomatic transmission, which had allowed it to spread around the world rapidly.2 This has fuelled concerns that this outbreak could evolve into a global pandemic.
As there are still several questions and uncertainty at this stage, there are bound to be concerns and anxiety among the general public towards the emerging situation. An important aspect of public health policy is designing and managing public communications. To do so, previous studies have found that social media analyses via Twitter are a feasible and novel method to study public sentiment and emotional manifestations on a given topic.3,4 Therefore, in this infodemiology study, we aimed to study the public sentiments on the emerging global outbreak of monkeypox and, in doing so, highlight and hopefully address the public's concerns.
Section snippets
Methods
Original tweets containing the terms ‘monkeypox’, ‘monkey pox’ or ‘monkey_pox’ and posted in English language from 6 May 2022 (first case detected in the United Kingdom) to 23 July 2022 were extracted. Retweets and duplicate tweets were excluded from study. Bidirectional Encoder Representations from Transformers (BERT) Named Entity Recognition5 was then applied to select individual users only. Topic modelling, specifically BERTopic,6 was used to generate coherent key concerns on the public
Results
A total of 1,028,326 initial tweets were identified in the period of 6 May to 23 July 2022. A flowchart illustrating the tweets selection process with the help of unsupervised machine learning technology was shown in Fig. 1.
BERTopic generated five topics related to the public discourse surrounding monkeypox, and the total prevalence of these five topics was 68.9%; the remaining 31.1% was from a topic that was omitted from the current results as the model generates a Miscellaneous topic that
Discussion
In this infodemiology study, we used unsupervised machine learning to analyse a large volume of free-text data from social media tweets and further categorised the arising broad themes through iterative thematic analysis. The public sentiments surrounding the global outbreak of monkeypox can be broadly demarcated into three themes: (1) concerns of safety, (2) stigmatisation of minority communities, and (3) lack of faith in public institutions.
First, the concerns of safety are expected,
Ethical approval
Ethical approval was not applicable. No human participants were involved.
Funding
This research did not receive any specific grant funding from agencies in the public, commercial or not for profit sectors.
Competing interests
None to declare.
Author contributions
T.M.L. conceived the original idea. Q.X.N., C.E.Y., Y.L.L., L.K.T.W. and T.M.L. carried out the study and the relevant data analysis and interpretation. All authors contributed to the data analysis and interpretation. All authors discussed the results, contributed to the writing of the
References (10)
- et al.
Outbreak of human monkeypox in Nigeria in 2017-18: a clinical and epidemiological report
Lancet Infect Dis
(2019) - et al.
The never-ending global emergence of viral zoonoses after COVID-19? The rising concern of monkeypox in Europe, North America and beyond
Trav Med Infect Dis
(2022) - et al.
Analyzing twitter data to evaluate people's attitudes towards public health policies and events in the era of COVID-19
Int J Environ Res Publ Health
(2021) - et al.
Examining the utility of social media in COVID-19 vaccination: unsupervised learning of 672,133 twitter posts
JMIR Public Health Surveill
(2021) - et al.
BERT: pre-training of deep bidirectional Transformers for language understanding
(2019)
Cited by (23)
Machine learning-based monkeypox virus image prognosis with feature selection and advanced statistical loss function
2024, Medicine in MicroecologySex, drugs and superbugs: The rise of drug resistant STIs
2023, SSM - Qualitative Research in HealthExpressing Uncertainty and Risk About the Mpox Outbreak: A Textual Analysis of Twitter Messaging
2024, Communication Studies