A hierarchical machine learning framework for the analysis of large scale animal movement data

Autores
Torney, Colin J.; Morales, Juan Manuel; Husmeier, Dirk
Año de publicación
2021
Idioma
inglés
Tipo de recurso
artículo
Estado
versión publicada
Descripción
Background: In recent years the field of movement ecology has been revolutionized by our ability to collect high-accuracy, fine scale telemetry data from individual animals and groups. This growth in our data collection capacity has led to the development of statistical techniques that integrate telemetry data with random walk models to infer key parameters of the movement dynamics. While much progress has been made in the use of these models, several challenges remain. Notably robust and scalable methods are required for quantifying parameter uncertainty, coping with intermittent location fixes, and analysing the very large volumes of data being generated. Methods: In this work we implement a novel approach to movement modelling through the use of multilevel Gaussian processes. The hierarchical structure of the method enables the inference of continuous latent behavioural states underlying movement processes. For efficient inference on large data sets, we approximate the full likelihood using trajectory segmentation and sample from posterior distributions using gradient-based Markov chain Monte Carlo methods. Results: While formally equivalent to many continuous-time movement models, our Gaussian process approach provides flexible, powerful models that can detect multiscale patterns and trends in movement trajectory data. We illustrate a further advantage to our approach in that inference can be performed using highly efficient, GPU-accelerated machine learning libraries. Conclusions: Multilevel Gaussian process models offer efficient inference for large-volume movement data sets, along with the fitting of complex flexible models. Applications of this approach include inferring the mean location of a migration route and quantifying significant changes, detecting diurnal activity patterns, or identifying the onset of directed persistent movements.
Fil: Torney, Colin J.. University of Glasgow; Reino Unido
Fil: Morales, Juan Manuel. University of Glasgow; Reino Unido. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Patagonia Norte. Instituto de Investigaciones en Biodiversidad y Medioambiente. Universidad Nacional del Comahue. Centro Regional Universidad Bariloche. Instituto de Investigaciones en Biodiversidad y Medioambiente; Argentina
Fil: Husmeier, Dirk. University of Glasgow; Reino Unido
Materia
ANIMAL MOVEMENT
LARGE-SCALE DATA
MACHINE LEARNING
Nivel de accesibilidad
acceso abierto
Condiciones de uso
https://creativecommons.org/licenses/by/2.5/ar/
Repositorio
CONICET Digital (CONICET)
Institución
Consejo Nacional de Investigaciones Científicas y Técnicas
OAI Identificador
oai:ri.conicet.gov.ar:11336/183698

id CONICETDig_c403efd39c1d2f4d211396d3c4108559
oai_identifier_str oai:ri.conicet.gov.ar:11336/183698
network_acronym_str CONICETDig
repository_id_str 3498
network_name_str CONICET Digital (CONICET)
spelling A hierarchical machine learning framework for the analysis of large scale animal movement dataTorney, Colin J.Morales, Juan ManuelHusmeier, DirkANIMAL MOVEMENTLARGE-SCALE DATAMACHINE LEARNINGhttps://purl.org/becyt/ford/1.6https://purl.org/becyt/ford/1Background: In recent years the field of movement ecology has been revolutionized by our ability to collect high-accuracy, fine scale telemetry data from individual animals and groups. This growth in our data collection capacity has led to the development of statistical techniques that integrate telemetry data with random walk models to infer key parameters of the movement dynamics. While much progress has been made in the use of these models, several challenges remain. Notably robust and scalable methods are required for quantifying parameter uncertainty, coping with intermittent location fixes, and analysing the very large volumes of data being generated. Methods: In this work we implement a novel approach to movement modelling through the use of multilevel Gaussian processes. The hierarchical structure of the method enables the inference of continuous latent behavioural states underlying movement processes. For efficient inference on large data sets, we approximate the full likelihood using trajectory segmentation and sample from posterior distributions using gradient-based Markov chain Monte Carlo methods. Results: While formally equivalent to many continuous-time movement models, our Gaussian process approach provides flexible, powerful models that can detect multiscale patterns and trends in movement trajectory data. We illustrate a further advantage to our approach in that inference can be performed using highly efficient, GPU-accelerated machine learning libraries. Conclusions: Multilevel Gaussian process models offer efficient inference for large-volume movement data sets, along with the fitting of complex flexible models. Applications of this approach include inferring the mean location of a migration route and quantifying significant changes, detecting diurnal activity patterns, or identifying the onset of directed persistent movements.Fil: Torney, Colin J.. University of Glasgow; Reino UnidoFil: Morales, Juan Manuel. University of Glasgow; Reino Unido. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Patagonia Norte. Instituto de Investigaciones en Biodiversidad y Medioambiente. Universidad Nacional del Comahue. Centro Regional Universidad Bariloche. Instituto de Investigaciones en Biodiversidad y Medioambiente; ArgentinaFil: Husmeier, Dirk. University of Glasgow; Reino UnidoBioMed Central2021-12info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionhttp://purl.org/coar/resource_type/c_6501info:ar-repo/semantics/articuloapplication/pdfapplication/pdfhttp://hdl.handle.net/11336/183698Torney, Colin J.; Morales, Juan Manuel; Husmeier, Dirk; A hierarchical machine learning framework for the analysis of large scale animal movement data; BioMed Central; Movement Ecology; 9; 1; 12-2021; 1-112051-3933CONICET DigitalCONICETenginfo:eu-repo/semantics/altIdentifier/url/https://movementecologyjournal.biomedcentral.com/articles/10.1186/s40462-021-00242-0info:eu-repo/semantics/altIdentifier/doi/10.1186/s40462-021-00242-0info:eu-repo/semantics/openAccesshttps://creativecommons.org/licenses/by/2.5/ar/reponame:CONICET Digital (CONICET)instname:Consejo Nacional de Investigaciones Científicas y Técnicas2025-09-29T09:41:40Zoai:ri.conicet.gov.ar:11336/183698instacron:CONICETInstitucionalhttp://ri.conicet.gov.ar/Organismo científico-tecnológicoNo correspondehttp://ri.conicet.gov.ar/oai/requestdasensio@conicet.gov.ar; lcarlino@conicet.gov.arArgentinaNo correspondeNo correspondeNo correspondeopendoar:34982025-09-29 09:41:40.622CONICET Digital (CONICET) - Consejo Nacional de Investigaciones Científicas y Técnicasfalse
dc.title.none.fl_str_mv A hierarchical machine learning framework for the analysis of large scale animal movement data
title A hierarchical machine learning framework for the analysis of large scale animal movement data
spellingShingle A hierarchical machine learning framework for the analysis of large scale animal movement data
Torney, Colin J.
ANIMAL MOVEMENT
LARGE-SCALE DATA
MACHINE LEARNING
title_short A hierarchical machine learning framework for the analysis of large scale animal movement data
title_full A hierarchical machine learning framework for the analysis of large scale animal movement data
title_fullStr A hierarchical machine learning framework for the analysis of large scale animal movement data
title_full_unstemmed A hierarchical machine learning framework for the analysis of large scale animal movement data
title_sort A hierarchical machine learning framework for the analysis of large scale animal movement data
dc.creator.none.fl_str_mv Torney, Colin J.
Morales, Juan Manuel
Husmeier, Dirk
author Torney, Colin J.
author_facet Torney, Colin J.
Morales, Juan Manuel
Husmeier, Dirk
author_role author
author2 Morales, Juan Manuel
Husmeier, Dirk
author2_role author
author
dc.subject.none.fl_str_mv ANIMAL MOVEMENT
LARGE-SCALE DATA
MACHINE LEARNING
topic ANIMAL MOVEMENT
LARGE-SCALE DATA
MACHINE LEARNING
purl_subject.fl_str_mv https://purl.org/becyt/ford/1.6
https://purl.org/becyt/ford/1
dc.description.none.fl_txt_mv Background: In recent years the field of movement ecology has been revolutionized by our ability to collect high-accuracy, fine scale telemetry data from individual animals and groups. This growth in our data collection capacity has led to the development of statistical techniques that integrate telemetry data with random walk models to infer key parameters of the movement dynamics. While much progress has been made in the use of these models, several challenges remain. Notably robust and scalable methods are required for quantifying parameter uncertainty, coping with intermittent location fixes, and analysing the very large volumes of data being generated. Methods: In this work we implement a novel approach to movement modelling through the use of multilevel Gaussian processes. The hierarchical structure of the method enables the inference of continuous latent behavioural states underlying movement processes. For efficient inference on large data sets, we approximate the full likelihood using trajectory segmentation and sample from posterior distributions using gradient-based Markov chain Monte Carlo methods. Results: While formally equivalent to many continuous-time movement models, our Gaussian process approach provides flexible, powerful models that can detect multiscale patterns and trends in movement trajectory data. We illustrate a further advantage to our approach in that inference can be performed using highly efficient, GPU-accelerated machine learning libraries. Conclusions: Multilevel Gaussian process models offer efficient inference for large-volume movement data sets, along with the fitting of complex flexible models. Applications of this approach include inferring the mean location of a migration route and quantifying significant changes, detecting diurnal activity patterns, or identifying the onset of directed persistent movements.
Fil: Torney, Colin J.. University of Glasgow; Reino Unido
Fil: Morales, Juan Manuel. University of Glasgow; Reino Unido. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Patagonia Norte. Instituto de Investigaciones en Biodiversidad y Medioambiente. Universidad Nacional del Comahue. Centro Regional Universidad Bariloche. Instituto de Investigaciones en Biodiversidad y Medioambiente; Argentina
Fil: Husmeier, Dirk. University of Glasgow; Reino Unido
description Background: In recent years the field of movement ecology has been revolutionized by our ability to collect high-accuracy, fine scale telemetry data from individual animals and groups. This growth in our data collection capacity has led to the development of statistical techniques that integrate telemetry data with random walk models to infer key parameters of the movement dynamics. While much progress has been made in the use of these models, several challenges remain. Notably robust and scalable methods are required for quantifying parameter uncertainty, coping with intermittent location fixes, and analysing the very large volumes of data being generated. Methods: In this work we implement a novel approach to movement modelling through the use of multilevel Gaussian processes. The hierarchical structure of the method enables the inference of continuous latent behavioural states underlying movement processes. For efficient inference on large data sets, we approximate the full likelihood using trajectory segmentation and sample from posterior distributions using gradient-based Markov chain Monte Carlo methods. Results: While formally equivalent to many continuous-time movement models, our Gaussian process approach provides flexible, powerful models that can detect multiscale patterns and trends in movement trajectory data. We illustrate a further advantage to our approach in that inference can be performed using highly efficient, GPU-accelerated machine learning libraries. Conclusions: Multilevel Gaussian process models offer efficient inference for large-volume movement data sets, along with the fitting of complex flexible models. Applications of this approach include inferring the mean location of a migration route and quantifying significant changes, detecting diurnal activity patterns, or identifying the onset of directed persistent movements.
publishDate 2021
dc.date.none.fl_str_mv 2021-12
dc.type.none.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion
http://purl.org/coar/resource_type/c_6501
info:ar-repo/semantics/articulo
format article
status_str publishedVersion
dc.identifier.none.fl_str_mv http://hdl.handle.net/11336/183698
Torney, Colin J.; Morales, Juan Manuel; Husmeier, Dirk; A hierarchical machine learning framework for the analysis of large scale animal movement data; BioMed Central; Movement Ecology; 9; 1; 12-2021; 1-11
2051-3933
CONICET Digital
CONICET
url http://hdl.handle.net/11336/183698
identifier_str_mv Torney, Colin J.; Morales, Juan Manuel; Husmeier, Dirk; A hierarchical machine learning framework for the analysis of large scale animal movement data; BioMed Central; Movement Ecology; 9; 1; 12-2021; 1-11
2051-3933
CONICET Digital
CONICET
dc.language.none.fl_str_mv eng
language eng
dc.relation.none.fl_str_mv info:eu-repo/semantics/altIdentifier/url/https://movementecologyjournal.biomedcentral.com/articles/10.1186/s40462-021-00242-0
info:eu-repo/semantics/altIdentifier/doi/10.1186/s40462-021-00242-0
dc.rights.none.fl_str_mv info:eu-repo/semantics/openAccess
https://creativecommons.org/licenses/by/2.5/ar/
eu_rights_str_mv openAccess
rights_invalid_str_mv https://creativecommons.org/licenses/by/2.5/ar/
dc.format.none.fl_str_mv application/pdf
application/pdf
dc.publisher.none.fl_str_mv BioMed Central
publisher.none.fl_str_mv BioMed Central
dc.source.none.fl_str_mv reponame:CONICET Digital (CONICET)
instname:Consejo Nacional de Investigaciones Científicas y Técnicas
reponame_str CONICET Digital (CONICET)
collection CONICET Digital (CONICET)
instname_str Consejo Nacional de Investigaciones Científicas y Técnicas
repository.name.fl_str_mv CONICET Digital (CONICET) - Consejo Nacional de Investigaciones Científicas y Técnicas
repository.mail.fl_str_mv dasensio@conicet.gov.ar; lcarlino@conicet.gov.ar
_version_ 1844613314757263360
score 13.070432