How consistent are citizen science data sources, an exploratory study using free automated image recognition apps for woody plant identification
Bibliographic record
Abstract
Rapid advances in artificial intelligence have led to an upsurge in automated image recognition phone apps. This has increased public involvement in the collection, identification (ID), and analysis of biological data. While this is good for the field of biological data monitoring and biodiversity conservation, it is not clear how consistent IDs are from different apps. The goal of this exploratory work is to verify the accuracy and consistency in plant species identification from two widely used and free apps, i.e., PlantNet and iNaturalist apps. This work was conducted by scanning leaf samples along Bruce Trail in the Niagara Escarpment Biosphere Reserve as well as from the Royal Botanical Gardens arboretum, both in Hamilton, Ontario. Results show over 90% consistency in the identification of woody plants at the level of genus. At the species level, the PlantNet app demonstrated 79% accuracy (i.e., 79 out of 100 species correctly identified), while the iNaturalist app demonstrated 44% accuracy. Enhancing species representation in the database for Southern Ontario might help particularly species in the families Betulaceae, Rosaceae, and Pinaceae. Complementary use of the apps is recommended as a cautionary measure to reduce the likelihood of error in species-level woody plant identification as well as using apps in conjunction with field guide.
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 | 0.052 | 0.177 |
| Meta-epidemiology (narrow) | 0.000 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.005 | 0.008 |
| Science and technology studies | 0.003 | 0.002 |
| Scholarly communication | 0.009 | 0.006 |
| Open science | 0.002 | 0.003 |
| 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".