O4E.1 Cross-walking countries’ industry classifications using concordance files compared to fuzzy data matching, to utilise an international exposure dataset
Bibliographic record
Abstract
Background Carcinogen exposure data can potentially guide the work of health and safety (H and S) regulators. This project aims to use CAREX Canada data to estimate carcinogen exposures in New Zealand industries. This requires the creation of a cross-walk between the countries’ industry classifications. Methods Agile and big-data-science methodologies were used to construct two versions of an industry classification cross-walk from the 2006 Australian and New Zealand Standard Industrial Classification (ANZSIC06) to the Canadian version of the 2002 North American Industrial Classification (NAICS2002), used by CAREX Canada. Firstly, concordance files from government statistics bureaus cross-walked the path: ANZSIC06 ->International Standard Industrial Classification of All Economic Activities Rev4 ->NAICS2017->NAICS2012->NAICS2007->NAICS2002. The cross-walk accounted for ‘one-to-many-to-one’, non-machine formats, and missing/erroneous values. Secondly, a fuzzy data matching pipeline was designed. Data preparation removed redundant, stop, and common domain words, and lemmatised using morphological analysis (e.g. fishing to fish). Data matching used a hybrid algorithm combining ‘JaroWinkler-distance’ and a token-sort approach (i.e. ignoring the positional occurrence of words in a sentence) to match descriptions. A trial-and-error approach was used to assign weightings and concatenate the hierarchical industry classification levels to improve match accuracy. Python language was used for implementation. For each method, random samples of 50 matches were manually classified as either poor or sufficient by two people. Disagreements were discussed and consensus reached. Results The concordance cross-walk sample had 52% (95% C.I. 38%–66%) sufficient matches compared to 84% (95% C.I. 74%–94%) for the fuzzy data matching pipeline cross-walk sample. Conclusions Cross-walking countries’ industry classifications using a fuzzy data matching pipeline was more accurate than using a concordance cross-walk. The pipeline is modular enough to easily include more components. This work is part of a vision to design a semantic big-data lake, enabling integration of any data relevant to H and S.
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 | 0.007 | 0.040 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.006 | 0.006 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.004 | 0.004 |
| Open science | 0.002 | 0.004 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.066 | 0.025 |
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".