Dataset Information

Love thy neighbour: automatic animal behavioural classification of acceleration data using the K-nearest neighbour algorithm.

ABSTRACT: Researchers hoping to elucidate the behaviour of species that aren't readily observed are able to do so using biotelemetry methods. Accelerometers in particular are proving particularly effective and have been used on terrestrial, aquatic and volant species with success. In the past, behavioural modes were detected in accelerometer data through manual inspection, but with developments in technology, modern accelerometers now record at frequencies that make this impractical. In light of this, some researchers have suggested the use of various machine learning approaches as a means to classify accelerometer data automatically. We feel uptake of this approach by the scientific community is inhibited for two reasons; 1) Most machine learning algorithms require selection of summary statistics which obscure the decision mechanisms by which classifications are arrived, and 2) they are difficult to implement without appreciable computational skill. We present a method which allows researchers to classify accelerometer data into behavioural classes automatically using a primitive machine learning algorithm, k-nearest neighbour (KNN). Raw acceleration data may be used in KNN without selection of summary statistics, and it is easily implemented using the freeware program R. The method is evaluated by detecting 5 behavioural modes in 8 species, with examples of quadrupedal, bipedal and volant species. Accuracy and Precision were found to be comparable with other, more complex methods. In order to assist in the application of this method, the script required to run KNN analysis in R is provided. We envisage that the KNN method may be coupled with methods for investigating animal position, such as GPS telemetry or dead-reckoning, in order to implement an integrated approach to movement ecology research.

SUBMITTER: Bidder OR

PROVIDER: S-EPMC3931648 | biostudies-literature | 2014

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Love thy neighbour: automatic animal behavioural classification of acceleration data using the K-nearest neighbour algorithm.

Bidder Owen R OR Campbell Hamish A HA Gómez-Laich Agustina A Urgé Patricia P Walker James J Cai Yuzhi Y Gao Lianli L Quintana Flavio F Wilson Rory P RP

PloS one 20140221 2

Researchers hoping to elucidate the behaviour of species that aren't readily observed are able to do so using biotelemetry methods. Accelerometers in particular are proving particularly effective and have been used on terrestrial, aquatic and volant species with success. In the past, behavioural modes were detected in accelerometer data through manual inspection, but with developments in technology, modern accelerometers now record at frequencies that make this impractical. In light of this, som ...[more]

PMID: 24586354

Similar Datasets

Project description:AimCompetition for food among populations of closely related species and conspecifics that occur in both sympatry and parapatry can be reduced by interspecific and intraspecific spatial segregation. According to predictions of niche partitioning, segregation is expected to occur at habitat boundaries among congeners and within habitats among conspecifics, while negative relationships in the density of species or populations will occur in areas of overlap. We tested these predictions by modelling the winter distributions of two crested penguin species from three colonies in the south-western Atlantic.LocationPenguins were tracked from two large colonies on the Falkland Islands and one in South Georgia, from where they dispersed through the South Atlantic, Southern Ocean and south-eastern Pacific.MethodsForty macaroni penguins (Eudyptes chrysolophus) from South Georgia and 82 southern rockhopper penguins (Eudyptes chrysocome chrysocome) from two colonies in the Falkland Islands were equipped with global location sensors which log time and light, allowing positions to be estimated twice-daily, from April to August in 2011. Positions were gridded and converted into maps of penguin density. Metrics of overlap were calculated and density was related to remote-sensed oceanographic variables and competitor density using generalized additive models.ResultsMacaroni penguins from western South Georgia and southern rockhopper penguins from Steeple Jason Island, Falkland Islands, were spatially segregated by differences in their habitat preferences thus supporting our first prediction regarding interspecific segregation. However, southern rockhopper penguins from Beauchêne Island showed a marked spatial overlap with macaroni penguins as the two had similar habitat preferences and strong mutual associations when controlling for habitat. Contrary to our predictions relating to intraspecific segregation, southern rockhopper penguins from Beauchêne Island and Steeple Jason Island were segregated by differences in habitat selection.Main conclusionsMorphological differentiation probably allows macaroni penguins from South Georgia and southern rockhopper penguins from Beauchêne Island to coexist in areas of spatial overlap, whereas segregation of the two Falkland rockhopper penguin populations may have arisen from two distinct lineages retaining cultural fidelity to ancestral wintering areas.

Dataset Information

Love thy neighbour: automatic animal behavioural classification of acceleration data using the K-nearest neighbour algorithm.

Publications

Love thy neighbour: automatic animal behavioural classification of acceleration data using the K-nearest neighbour algorithm.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets