Detection and classification of North Atlantic right whales in the bay of fundy using independent component analysis
Bibliographic record
Abstract
A novel m ethod of detection and classification for m arine m am m als is presented which uses tech niques from independent com ponent analysis to solve the blind source separation problem for N orth A tlantic right whales (Eubalaena glacialis).Using the fundam entally non-G aussian nature of m arine m am m al vocalizations and d a ta collected on m ultiple hydrophones, we are able to separate right whale source spectra, up to an unknown scale, from am bient noise.This technique assumes th a t the array d a ta is a linear com bination of non-G aussian source signals b u t does not require specific knowledge of the array geometry.A detection algorithm which separates right whale vocalizations from am bient background using a Kolmogorov-Smirnov test statistic is presented and tested on d a ta collected in the Bay of Fundy.The perform ance of the detector was found to be such th a t it was possible to achieve a probability of detection of about three-fourths w ith a false alarm probability of about one-third.Independent com ponent analysis was found to provide little improvement over standard principle com ponent analysis, which was used as preprocessing step. RSUMUne nouvelle m ethode de detection et de classification pour les mammifres m arins est presentee.Elle utilise des techniques d 'analyse par com posantes independantes pour rsoudre des problemes de separation aveugle de sources pour des baleines franches de l'A tlantique Nord (Eubalaena glacialis).E n se basant sur la n ature non gaussienne des vocalisations des mammiferes m arins et sur les donnees recueillies par un ensemble d 'hydrophones, nous avons ete capables de separer les spectres de baleines franches, ju sq u ' une echelle inconnue, du bruit am biant.C ette technique suppose que l 'ensemble des donnees est une combinaison lineaire des signaux sources non gaussiens, mais ne requiert pas de connaissance particuliere sur la geometrie de l 'ensemble des hydrophones.Un algorithm e de detection p erm ettan t de separer les vocalisations de baleines franches du bruit am biant en utilisant un test statistique Kolmogorov-Smirnov est prsent et teste sur des donnees recueillies dans la Baie de Fundy.La perform ance du detecteur etait telle q u 'il a ete possible de raliser une probabilit de detection d 'environ trois quarts, avec une probabilit de fausse alarm e d 'environ un tiers.L 'analyse par com posantes independantes n 'a donne que des am eliorations mineures com pare a l 'analyse par com posantes principales stan d ard qui a ete utilisee comme etape de pre-traitem ent.
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 | 0.000 | 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 | 0.000 | 0.000 |
| 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".