Digitally-assisted iconology: A method for the analysis of digital media
Bibliographic record
Abstract
Exploring medium-to-large datasets of social media imagery can be challenging. This paper describes a digitally-assisted iconology, a hybrid methodology that includes machine learning and data analytics for sorting through medium-sized datasets of images that lack metadata to describe their pictorial content. The method plays to the strengths of current digital technologies. Using machine learning, pictures are first clustered in a preliminary stage based upon basic formal presentational characteristics. Thematic analysis follows this preliminary stage, based upon an expansion of Aby Warburg’s “pre-coined expressive values”, which are frequently found in pictures displaying high levels of user reception. Once clustered via these two separate stages, the researcher can then drill down using familiar forms of visual analysis to explore how similar concepts have been rendered in different ways. The analysis may be augmented by exploring the commentary appended to these pictures, which adds a further level of detail providing insight into end-user interpretations. The approach – including its drawbacks – is demonstrated via a consequential dataset of pictures shared on Twitter in 2015, after a Syrian child was found drowned off the Turkish shore. Derivative imagery based upon the original photographs referenced longstanding iconographic themes.
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 | 0.004 | 0.014 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.014 | 0.008 |
| Science and technology studies | 0.002 | 0.003 |
| Scholarly communication | 0.006 | 0.005 |
| Open science | 0.001 | 0.004 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.017 | 0.005 |
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".