Automatic revision of rules used to guide the generalisation process in systems based on a trial and error strategy
Bibliographic record
Abstract
Automating the generalisation process, a major issue for national mapping agencies, is extremely complex. Several works have proposed to deal with this complexity using a trial and error strategy. The performance of systems based on such a strategy is directly dependent on the quality of the control knowledge (i.e. heuristics) used to guide the trials. Unfortunately, most of the time, the definition and updation of knowledge is a fastidious task. In this context, automatic knowledge revision can not only improve the performance of the generalisation, but also allow it to automatically adapt to various usages and evolve when new elements are introduced. In this article, an offline knowledge revision approach is proposed, based on a logging of the system and on the analysis of outcoming logs. This approach is dedicated to the revision of control knowledge expressed by production rules. We have implemented and tested this approach for the automated generalisation of groups of buildings within a generalisation model called AGENT, from initial data that reference a scale of approximately 1:15,000 compared with the target map's scale of 1:50,000. The results show that our approach improves the quality of the control knowledge and thus the performance of the system. Moreover, the approach proposed is generic and can be applied to other systems based on a trial and error strategy, dedicated to generalisation or not.
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 | 0.059 | 0.307 |
| Meta-epidemiology (narrow) | 0.003 | 0.002 |
| Meta-epidemiology (broad) | 0.005 | 0.003 |
| Bibliometrics | 0.006 | 0.003 |
| Science and technology studies | 0.002 | 0.004 |
| Scholarly communication | 0.009 | 0.008 |
| Open science | 0.008 | 0.007 |
| Research integrity | 0.005 | 0.006 |
| Insufficient payload (model declined to judge) | 0.006 | 0.003 |
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".