Interpreting the Inscrutable: Ethnographic Approaches to Studying the Development of Machine Learning Models
Bibliographic record
Abstract
The goal of this paper is to outline methodological insights and tips that organizational ethnographers can employ in studying the development of machine learning models (“ML tools”) to better understand the resulting ML tool and its organizational consequences. While proliferating at a swift pace across different organizations, ML tools do not easily lend themselves to be observed because they are dynamic and frequently changing. Recognizing this challenge, we propose focusing on data work as a way to capture the concrete traces that help make sense of the resulting ML tool’s intelligent functions which are often described as inscrutable. We draw on illustrative examples from our fieldwork experiences in two teaching hospitals in the Netherlands and China, where we observed the data work involved in creating ML tools. Along the three stages of data work performed during ML tool development: collecting, annotating, and recalibrating, we propose methodological insights and tips that can help ethnographers uncover how ML tools are shaped by relations among entities, both pre-existing and emerging, across levels of analysis.
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 | 0.026 | 0.034 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.004 | 0.004 |
| Science and technology studies | 0.007 | 0.027 |
| Scholarly communication | 0.009 | 0.014 |
| Open science | 0.003 | 0.008 |
| Research integrity | 0.002 | 0.005 |
| Insufficient payload (model declined to judge) | 0.003 | 0.001 |
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".