A national discussion of COVID-19 on Twitter
Bibliographic record
Abstract
Introduction: During the COVID-19 pandemic, social media became increasingly relied upon for health information in Canada. By analyzing georeferenced tweets using natural language processing, we aimed to understand regional discussions and concerns about school closures, masking, vaccines, and lockdowns during the pandemic's first two years. Methods: Using Twitter's application programming interface, we collected English-language tweets with keywords related to COVID-19 posted between January 1, 2020 and February 22, 2022 from Canadian users. Results: Out of all retained tweets, 2,851,951 (47.9%) were about vaccines, 1,344,008 (22.6%) about lockdowns, 1,011,909 (17%) about schooling, and 752,014 (12.6%) about masking. Tweets on schooling received the most engagement, with the highest rates of likes (17.3%), retweets (18.7%), replies (10%), and quotes (6.8%). The most common emotions expressed were trust, fear, and anticipation, with lockdown tweets showing greater fear and sadness. Overall, sentiment was negative, particularly regarding lockdowns in the Northwest Territories and Alberta. Discussion: During the COVID-19 pandemic, Twitter became an essential tool for analyzing public sentiment regarding government actions. Users showed the most interest in vaccines, followed by lockdowns, schooling, and masking, with the highest engagement on schooling tweets. Our analysis of sentiment, emotion, and content revealed valuable insights into public beliefs about COVID-19 in Canada, highlighting regional differences and shifts in sentiment, particularly negative reactions to school closures as government recommendations evolved. Our study adds to the growing evidence supporting the use of natural language processing for real-time analysis of social media content to early identify public health concerns.
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 | 0.003 | 0.008 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.012 | 0.002 |
| Scholarly communication | 0.004 | 0.003 |
| Open science | 0.001 | 0.003 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.007 | 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".