Unmasking the Twitter Discourses on Masks During the COVID-19 Pandemic: User Cluster–Based BERT Topic Modeling Approach
Bibliographic record
Abstract
Background: The COVID-19 pandemic has spotlighted the politicization of public health issues. A public health monitoring tool must be equipped to reveal a public health measure's political context and guide better interventions. In its current form, infoveillance tends to neglect identity and interest-based users, hence being limited in exposing how public health discourse varies by different political groups. Adopting an algorithmic tool to classify users and their short social media texts might remedy that limitation. Objective: We aimed to implement a new computational framework to investigate discourses and temporal changes in topics unique to different user clusters. The framework was developed to contextualize how web-based public health discourse varies by identity and interest-based user clusters. We used masks and mask wearing during the early stage of the COVID-19 pandemic in the English-speaking world as a case study to illustrate the application of the framework. Methods: We first clustered Twitter users based on their identities and interests as expressed through Twitter bio pages. Exploratory text network analysis reveals salient political, social, and professional identities of various user clusters. It then uses BERT Topic modeling to identify topics by the user clusters. It reveals how web-based discourse has shifted over time and varied by 4 user clusters: conservative, progressive, general public, and public health professionals. Results: This study demonstrated the importance of a priori user classification and longitudinal topical trends in understanding the political context of web-based public health discourse. The framework reveals that the political groups and the general public focused on the science of mask wearing and the partisan politics of mask policies. A populist discourse that pits citizens against elites and institutions was identified in some tweets. Politicians (such as Donald Trump) and geopolitical tensions with China were found to drive the discourse. It also shows limited participation of public health professionals compared with other users. Conclusions: We conclude by discussing the importance of a priori user classification in analyzing web-based discourse and illustrating the fit of BERT Topic modeling in identifying contextualized topics in short social media texts.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.005 | 0.009 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.003 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".