Utilising urban context recognition and machine learning to improve the generalisation of buildings
Bibliographic record
Abstract
The introduction of automated generalisation procedures in map production systems requires that generalisation systems are capable of processing large amounts of map data in acceptable time and that cartographic quality is similar to traditional map products. With respect to these requirements, we examine two complementary approaches that should improve generalisation systems currently in use by national topographic mapping agencies. Our focus is particularly on self‐evaluating systems, taking as an example those systems that build on the multi‐agent paradigm. The first approach aims to improve the cartographic quality by utilising cartographic expert knowledge relating to spatial context. More specifically, we introduce expert rules for the selection of generalisation operations based on a classification of buildings into five urban structure types, including inner city, urban, suburban, rural, and industrial and commercial areas. The second approach aims to utilise machine learning techniques to extract heuristics that allow us to reduce the search space and hence the time in which a good cartographical solution is reached. Both approaches are tested individually and in combination for the generalisation of buildings from map scale 1:5000 to the target map scale of 1:25 000. Our experiments show improvements in terms of efficiency and effectiveness. We provide evidence that both approaches complement each other and that a combination of expert and machine learnt rules give better results than the individual approaches. Both approaches are sufficiently general to be applicable to other forms of self‐evaluating, constraint‐based systems than multi‐agent systems, and to other feature classes than buildings. Problems have been identified resulting from difficulties to formalise cartographic quality by means of constraints for the control of the generalisation process.
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 | 0.002 | 0.007 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
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".