Capturing expert knowledge for ecosystem mapping using Bayesian networks
Bibliographic record
Abstract
Large-scale ecosystem maps are essential tools for managers of forest-related activities. In British Columbia, the prevailing approach for ecosystem mapping has been to use an expert system that captures expert knowledge in the form of a belief matrix. In this project, a Bayesian network rather than a belief matrix was used in an attempt to overcome some of the drawbacks of the belief-matrix approach. A Bayesian-network knowledge base was created for each of the following three biogeoclimatic variants: montane very wet maritime coastal western hemlock (CWHvm2), submontane very wet maritime coastal western hemlock (CWHvm1), and central very wet hypermaritime coastal western hemlock (CWHvh2), and applied to a study area encompassing Prince Rupert. A map of ecosystems by grouping site series was produced using each of the knowledge bases. Accuracy assessments performed on each of the maps of grouped site series revealed that the maps poorly predicted the spatial distribution of uncommon and very wet site-series groups. For example, overall map accuracy for the CWHvm2, CWHvm1, and CWHvh2 variants was 47.8%, 50.3%, and 33.3%, respectively. The results of the map-accuracy assessment, however, were consistent with those resulting from a belief-matrix approach conducted in an earlier study. We feel that Bayesian network knowledge bases are easier to develop, interpret, and update than belief matrices.
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 | 0.009 | 0.053 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.006 | 0.003 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.005 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.003 | 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".