Mission: Implausible — Revealing rogue marine species in records across biodiversity data platforms
Bibliographic record
Abstract
Online biodiversity platforms publish datasets with graphic tools to help with quality control of submitted records, but more could be done to make the data robust for ecological analyses. Attention has focused mostly on automating tools for obvious errors, including misspelled names and synonyms, dates, or coordinates. However, a manual review of species identifications and distributions may uncover improbable records, such as a species reported in an area far from its usual range, or a rare species found in an area that has many more records of a related species. Examples are shown by constructing checklists in the Northwest Atlantic, using information from the World Register of Marine Species (WoRMS, http://www.marinespecies.org) and the Ocean Biogeographic Information System (OBIS, https://obis.org). Reviewing rare species records revealed some misidentifications, but in other instances, the rare species was valid while it was the commonly reported species that needed correction. Confirmations were obtained by comparing records from different regions, but also across platforms, including photos from observers on iNaturalist Canada (https://inaturalist.ca), genetic analyses on Barcode of Life Data systems (BOLD, http://www.boldsystems.org), and literature in the Biodiversity Heritage Library (BHL, https://www.biodiversitylibrary.org). While this exercise succeeded in validating the marine taxa of a region, it is an obvious candidate for automation in three areas: 1) flagging records of improbable taxa in a region, 2) comparing records with different types of information (e.g., specimen photos, genetic groupings, or literature records), and 3) updating users and providers when records get flagged as unusual or are modified. The first approach could be explored using online graphics tools or R software packages (rOpenSci, https://ropensci.org). The second toolset, comparing records across platforms, is partially realized with some linkages already operating between WoRMS, OBIS, BOLD, BHL, iNaturalist, and the Global Biodiversity Information Facility (GBIF, https://www.gbif.org). The third target will be the most difficult to implement, requiring reliable platform cross-linkages and specimen record identifiers to send notifications of changed status of records to both users and the original data source. Ongoing work is discussed on communicating the need to review records across platforms, with the hope that toolsets will be developed to make this task easier.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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 teacher head, 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".