Can Consumer-Posted Photos Serve as a Leading Indicator of Restaurant Survival? Evidence from Yelp
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Despite the substantial economic impact of the restaurant industry, large-scale empirical research on restaurant survival has been sparse. We investigate whether consumer-posted photos can serve as a leading indicator of restaurant survival above and beyond reviews, firm characteristics, competitive landscape, and macroconditions. We employ machine learning techniques to extract features from 755,758 photos and 1,121,069 reviews posted on Yelp between 2004 and 2015 for 17,719 U.S. restaurants. We also collect data on restaurant characteristics (e.g., cuisine type, price level) and competitive landscape as well as entry and exit (if applicable) time from each restaurant’s Yelp/Facebook page, own website, or Google search engine. Using a predictive XGBoost algorithm, we find that consumer-posted photos are strong predictors of restaurant survival. Interestingly, the informativeness of photos (e.g., the proportion of food photos) relates more to restaurant survival than do photographic attributes (e.g., composition, brightness). Additionally, photos carry more predictive power for independent, young or mid-aged, and medium-priced restaurants. Assuming that restaurant owners possess no knowledge about future photos and reviews, photos can predict restaurant survival for up to three years, whereas reviews are only predictive for one year. We further employ causal forests to facilitate the interpretation of our predictive results. Among photo content variables, the proportion of food photos has the largest positive association with restaurant survival, followed by proportions of outside and interior photos. Among others, the proportion of photos with helpful votes also positively relates to restaurant survival. This paper was accepted by Juanjuan Zhang, marketing. Funding: The authors thankNvidia and Clarifai for supporting this research. Supplemental Material: The online appendix and data are available at https://doi.org/10.1287/mnsc.2022.4359 .
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.
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 | 0.005 | 0.002 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.002 |
| Science and technology studies | 0.002 | 0.001 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 it