Selective and Automatic editing with CADI-applications
Bibliographic record
Abstract
When designing a CADI-machine for a business survey, the implicit approach with Blaise is to edit every record with the same effort, even though some firms contribute vastly more to publication figures than others. Canadian, Swedish and Dutch research into the effectiveness of this approach has shown that in the cases under investigation at least 50 % of the edits have virtually no effect on publication figures. This calls for methods to skip ineffective editing activities. One way is to select records which have a high risk of containing influential errors, the critical stream, for Blaise-editing. The remaining records, the non-critical stream, may remain unedited. The part of the non-critical stream that has Blaise status 'dirty' or 'suspect ' may also be 'cleaned' by a routine for automatic editing. The art of selective editing is to devise a powerful and yet practical formula to determine the risk that a record contains influential errors. This involves taking account of inclusion probabilities, non-response probabilities, size of the publication cell and, most important of all, a benchmark to determine whether an observed score may be in error, like the cell-mean (or median) for that variable. Partly drawing on experiences from official statistics in other countries like Canada, the US and Sweden, the principles of automatic editing will be briefly dealt with.
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 | 0.032 | 0.210 |
| Meta-epidemiology (narrow) | 0.002 | 0.002 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.007 | 0.005 |
| Science and technology studies | 0.002 | 0.002 |
| Scholarly communication | 0.008 | 0.005 |
| Open science | 0.004 | 0.006 |
| Research integrity | 0.002 | 0.003 |
| Insufficient payload (model declined to judge) | 0.046 | 0.031 |
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".