Heritage site-seeing through the visitor’s lens on Instagram
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
English Heritage is a charity that manages over 400 historic sites in the UK, from prehistoric sites to medieval castles, most of them free, non-ticketed, and unstaffed. As such, there is little information about visitor attendance and behaviour in those sites—a challenge common to other non-ticketed heritage sites. In this context, image-based social media such as Instagram appear as a possible solution, as photographs are often central to the tourist experience, and tourists present their imagined audiences with a self-narrative of their trip. Therefore, this study aims to improve our understanding of tourist behaviour in unstaffed heritage sites by analysing publicly available Instagram data. We collect posts on unstaffed English Heritage sites, finding that posting activity concentrates at a few sites. Focusing on 3,979 images each for the top five sites, we analyse image content using pre-trained object detection models. Besides off-the-shelf inference, we fine-tune a model to identify structures from particular heritage sites, and are able to describe the types of photographs taken by visitors in each site, supporting the notion of tourists as performers with the site serving as backdrop. Overall, this study demonstrates a methodology for understanding cultural behaviour at heritage sites using images from social media posts. In addition to recovering the otherwise lost connection between a heritage organisation and its visitors, our methodology can be readily extended to other tourist destinations to understand how visitors interact with and relate to these sites and the objects within them through their photographs.
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 | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.002 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.000 |
| 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 it