Microplastic Spectral Classification Needs an Open Source Community: Open Specy to the Rescue!
Bibliographic record
Abstract
Microplastic pollution research has suffered from inadequate data and tools for spectral (Raman and infrared) classification. Spectral matching tools often are not accurate for microplastics identification and are cost-prohibitive. Lack of accuracy stems from the diversity of microplastic pollutants, which are not represented in spectral libraries. Here, we propose a viable software solution: Open Specy. Open Specy is on the web (www.openspecy.org) and in an R package. Open Specy is free and allows users to view, process, identify, and share their spectra to a community library. Users can upload and process their spectra using smoothing (Savitzky-Golay filter) and polynomial baseline correction techniques (IModPolyFit). The processed spectrum can be downloaded to be used in other applications or identified using an onboard reference library and correlation-based matching criteria. Open Specy's data sharing and session log features ensure reproducible results. Open Specy houses a growing library of reference spectra, which increasingly represents the diversity of microplastics as a contaminant suite. We compared the functionality and accuracy of Open Specy for microplastic identification to commonly used spectral analysis software. We found that Open Specy was the only open source software and the only software with a community library, and Open Specy had comparable accuracy to popular software (OMNIC Picta and KnowItAll). Future developments will enhance spectral identification accuracy as the reference library and functionality grows through community-contributed spectra and community-developed code. Open Specy can also be used for applications beyond microplastic analysis. Open Specy's source code is open source (CC-BY-4.0, attribution only) (https://github.com/wincowgerDEV/OpenSpecy).
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 | 0.012 | 0.045 |
| Meta-epidemiology (narrow) | 0.004 | 0.002 |
| Meta-epidemiology (broad) | 0.003 | 0.004 |
| Bibliometrics | 0.007 | 0.006 |
| Science and technology studies | 0.003 | 0.002 |
| Scholarly communication | 0.007 | 0.017 |
| Open science | 0.006 | 0.013 |
| Research integrity | 0.003 | 0.006 |
| Insufficient payload (model declined to judge) | 0.056 | 0.093 |
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".