Image auto-coding tools for social impact assessment: leveraging social media data to understand human dimensions of hydroelectricity landscape changes in Canada
Bibliographic record
Abstract
Social media data has been shown to be a valuable data source for assessing social impacts, particularly when paired with the swift progress in artificial intelligence technologies, allowing comprehensive analyses of larger datasets than is possible using conventional approaches. We sought to understand the social impacts of hydropower-related landscape changes based on a quasi-chronosequence of three study cases in Canada, using social media images in conjunction with machine learning to conduct image and textual analysis. We employed the Google Cloud Vision API, a pre-trained artificial intelligence (AI)-based tool, to detect labels from over 19,000 landscape images of the relevant regions sourced from Instagram. This yielded a comprehensive set of over 188,000 labels. We used a generative probabilistic model (Latent Dirichlet allocation, an unsupervised machine learning algorithm) to create clusters based on the labels. These clusters revealed prevalent landscape features, human activities, and animate and inanimate objects—as well as which frequently co-occurred—allowing us to understand and predict some of the social impacts of the landscape changes potentially caused by hydroelectric dams and reservoirs. This provides an example of integrating social media data and automated analysis tools powered by machine/deep learning into social impact assessment. Notably, such pre-trained (or “ready-to-use”) tools require minimum programming skills, benefiting scholars and practitioners who are less versed in technical domains. The insights gained from hydroelectricity case studies can also inform decisions about energy transition.
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 | 0.000 | 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 | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| 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".