Format

Send to

Choose Destination
J Exp Biol. 2018 Apr 16. pii: jeb.177378. doi: 10.1242/jeb.177378. [Epub ahead of print]

Combined use of two supervised learning algorithms to model sea turtle behaviours from tri-axial acceleration data.

Author information

1
Centre National de la Recherche Scientifique, Institut Pluridisciplinaire Hubert Curien, UMR 7178 CNRS-Université Louis Pasteur, Département d'Ecologie, Physiologie et Ethologie, Strasbourg, France lorene.jeantet@iphc.cnrs.fr.
2
Aquarium La Rochelle, La Rochelle, France.
3
Centre National de la Recherche Scientifique, Institut Pluridisciplinaire Hubert Curien, UMR 7178 CNRS-Université Louis Pasteur, Département d'Ecologie, Physiologie et Ethologie, Strasbourg, France.
4
Direction de l'Environnement, de l'Aménagement et du Logement Martinique, Schoelcher, Martinique.
5
Office de l'Eau Martinique, Fort-de-France, Martinique.
6
Institut Pasteur de la Guyane, Cayenne, Guyane.
7
Department of Biosciences, College of Science, Swansea University, Swansea, UK.

Abstract

Accelerometers are becoming ever more important sensors in animal-attached technology, providing data that allow determination of body posture and movement and thereby helping elucidate behaviour in animals that are difficult to observe.We sought to validate the identification of sea turtle behaviours from accelerometer signals by deploying tags on the carapace of a juvenile loggerhead (Caretta caretta), an adult hawksbill (Eretmochelys imbricata) and an adult green turtle (Chelonia mydas) at Aquarium La Rochelle. We recorded tri-axial acceleration at 50Hz for each species for a full day while two fixed cameras recorded their behaviours. We identified behaviours from the acceleration data using two different supervised learning algorithms; Random Forest and Classification And Regression Tree (CART), treating the data from the adult animals as separate from the juvenile data. We achieved a global accuracy of 81.30% for the adult turtle CART model and 71.63% for the juvenile loggerhead, identifying 10 and 12 different behaviours, respectively. Equivalent figures were 86.96% for the hawksbill and green turtle Random Forest model and 79.49% for the loggerhead, for the same behaviours. The use of Random Forest combined with CART algorithms allowed us to understand the decision rules implicated in behaviour discrimination, and thus remove or group together some "confused" or underrepresented behaviours in order to get the most accurate models. This study is the first to validate accelerometer data to identify turtle behaviours and the approach can now be tested on other captive sea turtle species.

KEYWORDS:

Accelerometry; Endangered species; Supervised learning algorithms

PMID:
29661804
DOI:
10.1242/jeb.177378

Supplemental Content

Full text links

Icon for HighWire
Loading ...
Support Center