O-174 Formation of the international partnership on automatic occupation coding – call for partners and collaboration
Bibliographic record
Abstract
Introduction Job coding is important for occupational epidemiology. Occupational classifications, such as the ILO’s International Standard Classification of Occupations (ISCO), are often used in job-exposure matrices (JEMs) and other models for exposure assessment in population-based studies. In these studies, assignment of job codes is often performed manually. This work is labourious, costly, and limited in reliability. Tools for automatic assignment of job codes are available for select coding systems and languages; however, their application in occupational epidemiology is limited mainly due to uncertainties around tool performance and how their use might impact exposure assessment. Material and Methods Following discussions held during and after EPICOH 2021, the International Partnership on Automatic Occupation Coding (IPAOC) was formed by a group of occupational exposure assessment scientists and epidemiologists. Aiming to promote knowledge sharing and collaborations on the development of automatic coding algorithms and software, IPAOC met regularly and actively sought new partners in 2022 while defining its research agenda. Results and Conclusions As of November 2022, IPAOC includes more than 40 members from six countries. The partnership is diverse and multidisciplinary; research areas represented include computer and data science, labour economics, occupational medicine, occupational health, official statistics, statistics, and sociology. Member interests in automatic job coding also span across a number of languages and occupation classifications systems, including in English (Coding: ISCO, US SOC and Canadian NOC), French (PCS), German (KldB), and Dutch (ISCO). For 2023, IPAOC’s goals are to address two main challenges for developing better automatic job coding tools: siloed development in separate projects/countries and low training data availability. Specifically, IPAOC will 1) apply for funding for a week-long workshop meeting to facilitate knowledge sharing and cooperation in the Lorentz Center in Leiden, the Netherlands; and 2) develop a shared benchmarking dataset for coding algorithm development.
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 | 0.089 | 0.094 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.004 | 0.002 |
| Science and technology studies | 0.006 | 0.004 |
| Scholarly communication | 0.009 | 0.009 |
| Open science | 0.004 | 0.030 |
| Research integrity | 0.008 | 0.009 |
| Insufficient payload (model declined to judge) | 0.067 | 0.035 |
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".