A deep neural network approach for sentiment analysis of medically related texts: an analysis of tweets related to concussions in sports
Bibliographic record
Abstract
Annually, over three million people in North America suffer concussions. Every age group is susceptible to concussion, but youth involved in sporting activities are particularly vulnerable, with about 6% of all youth suffering a concussion annually. Youth who suffer concussion have also been shown to have higher rates of suicidal ideation, substance and alcohol use, and violent behaviors. A significant body of research over the last decade has led to changes in policies and laws intended to reduce the incidence and burden of concussions. However, it is also clear that youth engaging in high-risk activities like sport often underreport concussion, while others may embellish reports for specific purposes. For such policies and laws to work, they must operate effectively within a facilitative social context so understanding the culture around concussion becomes essential to reducing concussion and its consequences. We present an automated deep neural network approach to analyze tweets with sport-related concussion context to identify the general public's sentiment towards concerns in sport-related concussion. A single-layer and multi-layer convolutional neural networks, Long Short-Term Memory (LSTM) networks, and Bidirectional LSTM were trained to classify the sentiments of the tweets. Afterwards, we train an ensemble model to aggregate the predictions of our networks to provide a final decision of the tweet's sentiment. The system achieves an evaluation F1 score of 62.71% based on Precision and Recall. The trained system is then used to analyze the tweets in the FIFA World Cup 2018 to measure audience reaction to events involving concussion. The neural network system provides an understanding of the culture around concussion through sentiment analysis.
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 | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.000 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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".