Machine Learning for Outlier Detection in Algal and Cyanobacterial Fluorescence Signals
Bibliographic record
Abstract
Many drinking water utilities drawing from waters susceptible to harmful algal blooms (HABs) are implementing monitoring tools that can alert them of the onset of potential blooms. Some have invested in fluorescence-based online monitoring probes to measure chlorophyll a and phycocyanin, two pigments found in cyanobacteria, but it is not clear how to best use the data generated this way. Previous studies have focused on correlating phycocyanin fluorescence and cyanobacteria cell counts. However, not all utilities collect cell count data, making this method impossible to apply in some cases. Instead, this paper proposes a novel approach to determine when a utility needs to respond to an HAB based on machine learning by identifying outliers in chlorophyll a and phycocyanin fluorescence data without the need for corresponding cell counts or biovolume. Four existing algorithms are evaluated on data collected at four buoys in Lake Erie from 2014-2019: k-means clustering, One-Class Support Vector Machine (SVM), elliptic envelope, and Isolation Forest (iForest). When trained and tested on data collected at different buoys, the iForest algorithm performed the best in terms of computation time for training and true positive rate, and second best for false positive rate. In a more realistic application where the algorithms are trained on historical phycocyanin data collected at the same location as the testing data, all the algorithms, except k-means, accurately identified anomalies in phycocyanin data coinciding with real cyanobacteria bloom events. Therefore, One-Class SVM, elliptic envelope, and iForest are promising algorithms for detecting potential HABs using fluorescence data.
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 | 0.004 | 0.009 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.002 | 0.001 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| 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 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".