Referee report. For: PUblications Metadata Augmentation (PUMA) pipeline [version 2; peer review: 1 approved, 1 approved with reservations]
Bibliographic record
Abstract
Cohort studies collect, generate and distribute data over long periods of time – often over the lifecourse of their participants. It is common for these studies to host a list of publications (which can number many thousands) on their website to demonstrate the impact of the study and facilitate the search of existing research to which the study data has contributed. The ability to search and explore these publication lists varies greatly between studies. We believe a lack of rich search and exploration functionality of study publications is a barrier to entry for new or prospective users of a study’s data, since it may be difficult to find and evaluate previous work in a given area. These lists of publications are also typically manually curated, resulting in a lack of rich metadata to analyse, making bibliometric analysis difficult. We present here a software pipeline that aggregates metadata from a variety of third-party providers to power a web based search and exploration tool for lists of publications. Alongside core publication metadata (i.e. author lists, keywords etc.), we include geocoding of first authors and citation counts in our pipeline. This allows a characterisation of a study as a whole based on common locations of authors, frequency of keywords, citation profile etc. This enriched publications metadata can be useful for generating study impact metrics and web-based graphics for public dissemination. In addition, the pipeline produces a research data set for bibliometric analysis or social studies of science. We use a previously published list of publications from a cohort study as an exemplar input data set to show the output and utility of the pipeline here.
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 | 0.013 | 0.146 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.006 | 0.006 |
| Science and technology studies | 0.004 | 0.001 |
| Scholarly communication | 0.008 | 0.006 |
| Open science | 0.003 | 0.005 |
| Research integrity | 0.005 | 0.003 |
| Insufficient payload (model declined to judge) | 0.789 | 0.651 |
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".