Online Influence and Sentiment of Fitness Tweets: Analysis of Two Million Fitness Tweets
Bibliographic record
Abstract
BACKGROUND: Publicly available fitness tweets may provide useful and in-depth insights into the real-time sentiment of a person's physical activity and provide motivation to others through online influence. OBJECTIVE: The goal of this experimental approach using the fitness Twitter dataset is two-fold: (1) to determine if there is a correlation between the type of activity tweet (either workout or workout+, which contains the same information as a workout tweet but has additional user-generated information), gender, and one's online influence as measured by Klout Score and (2) to examine the sentiment of the activity-coded fitness tweets by looking at real-time shared thoughts via Twitter regarding their experiences with physical activity and the associated mobile fitness app. METHODS: The fitness tweet dataset includes demographic and activity data points, including minutes of activity, Klout Score, classification of each fitness tweet, the first name of each fitness tweet user, and the tweet itself. Gender for each fitness tweet user was determined by a first name comparison with the US Social Security Administration database of first names and gender. RESULTS: Over 184 days, 2,856,534 tweets were collected in 23 different languages. However, for the purposes of this study, only the English-language tweets were analyzed from the activity tweets, resulting in a total of 583,252 tweets. After assigning gender to Twitter usernames based on the Social Security Administration database of first names, analysis of minutes of activity by both gender and Klout influence was determined. The mean Klout Score for those who shared their workout data from within four mobile apps was 20.50 (13.78 SD), less than the general Klout Score mean of 40, as was the Klout Score at the 95th percentile (40 vs 63). As Klout Score increased, there was a decrease in the number of overall workout+ tweets. With regards to sentiment, fitness-related tweets identified as workout+ reflected a positive sentiment toward physical activity by a ratio of 4 to 1. CONCLUSIONS: The results of this research suggest that the users of mobile fitness apps who share their workouts via Twitter have a lower Klout Score than the general Twitter user and that users who chose to share additional insights into their workouts are more positive in sentiment than negative. We present a novel perspective into the physical activity messaging from within mobile fitness apps that are then shared over Twitter. By moving beyond the numbers and evaluating both the Twitter user and the emotions tied to physical activity, future research could analyze additional relationships between the user's online influence, the enjoyment of the physical activity, and with additional analysis a long-term retention strategy for the use of a fitness app.
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 | 0.000 | 0.002 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.000 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| 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".