Categorizing Vaccine Confidence With a Transformer-Based Machine Learning Model: Analysis of Nuances of Vaccine Sentiment in Twitter Discourse
Bibliographic record
Abstract
BACKGROUND: Social media has become an established platform for individuals to discuss and debate various subjects, including vaccination. With growing conversations on the web and less than desired maternal vaccination uptake rates, these conversations could provide useful insights to inform future interventions. However, owing to the volume of web-based posts, manual annotation and analysis are difficult and time consuming. Automated processes for this type of analysis, such as natural language processing, have faced challenges in extracting complex stances such as attitudes toward vaccination from large amounts of text. OBJECTIVE: The aim of this study is to build upon recent advances in transposer-based machine learning methods and test whether transformer-based machine learning could be used as a tool to assess the stance expressed in social media posts toward vaccination during pregnancy. METHODS: A total of 16,604 tweets posted between November 1, 2018, and April 30, 2019, were selected using keyword searches related to maternal vaccination. After excluding irrelevant tweets, the remaining tweets were coded by 3 individual researchers into the categories Promotional, Discouraging, Ambiguous, and Neutral or No Stance. After creating a final data set of 2722 unique tweets, multiple machine learning techniques were trained on a part of this data set and then tested and compared with the human annotators. RESULTS: We found the accuracy of the machine learning techniques to be 81.8% (F score=0.78) compared with the agreed score among the 3 annotators. For comparison, the accuracies of the individual annotators compared with the final score were 83.3%, 77.9%, and 77.5%. CONCLUSIONS: This study demonstrates that we are able to achieve close to the same accuracy in categorizing tweets using our machine learning models as could be expected from a single human coder. The potential to use this automated process, which is reliable and accurate, could free valuable time and resources for conducting this analysis, in addition to informing potentially effective and necessary interventions.
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 | 0.017 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.001 | 0.001 |
| 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".