From data to insights: Machine learning in thematic analysis of complex health conditions on social media
Bibliographic record
Abstract
Social media data has the potential to enable the exploration of public perspectives on health conditions, interventions and policies. However, the resource-intensive nature of qualitative analysis creates a barrier to the timely utilization of social media data. Artificial intelligence can provide innovative ways to reduce the burden by augmenting the data analysis process for researchers. Therefore, the purpose of this research is to explore the feasibility of using Machine Learning (ML) in augmenting thematic analysis of complex health issues data available on social media. First, we performed a human-determined deductive thematic analysis of the 7,177 comments posted by YouTube video viewers on postpartum depression. Then we used the same data to perform machine-assisted analysis using five Natural Language Processing (NLP) classification models with and without class balancing techniques (class weight balance and SMOTE) to balance the unequal number of comments across themes. Our analysis suggested that supervised machine learning may not be optimal for analyzing complex health datasets. However, integrating ML-NLP techniques with thematic analysis can effectively filter out ineligible data, thereby enhancing the efficiency of traditional thematic analysis processes. This integration allows researchers to focus on valuable data, producing meaningful insights and comprehensive analysis while saving time otherwise spent sifting through a large amount of ineligible data. Social media data can offer significant public health insights, and to enhance the use of such data, future research should focus on developing artificial intelligence tools to improve the efficiency of thematic analysis using larger datasets and unsupervised machine learning models for analyzing complex health issues.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
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 teacher head, 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".