Text-mining metadata: What can titles tell us of the history of modern and contemporary art?
Bibliographic record
Abstract
The use of statistical text-mining to investigate the linguistic structure of textual resources has received limited attention in digital art history. In this paper the question I address is that of what text-mining titles as given in metadata can tell us about the history of modern and contemporary art. To investigate this question I constructed a dataset from the metadata for over 170,000 artworks, drawing on the online collections of 133 art museums in 30 countries. The use of topic modelling, parts-of-speech tagging and word counting allows me to identify large-scale and long-run patterns in the language used in the titles of those artworks. I set out an art historical reading of those patterns in which artistic interests signalled by the language used in titles come and go and are re-inflected, epistemic perspectives on the kinds of knowledge art can or should engender change, and artists engage with the ways that the title functions. My ‘distant’ reading is consistent with the canonical history of modern and contemporary art and cuts across the particularities of artist, period or movement which often feature in such accounts, providing a fresh perspective on that history. It also complements and extends the scholarship on the history of titles in the visual arts. The analytical framework and the dataset I have developed are not limited to answering the question addressed in this paper, and I consider some of the possibilities for future work.
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 | 0.006 | 0.040 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.014 | 0.017 |
| Science and technology studies | 0.002 | 0.002 |
| Scholarly communication | 0.005 | 0.015 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 0.002 |
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".