Love thy neighbour

Automatic animal behavioural classification of acceleration data using the k-nearest neighbour algorithm

Owen R Bidder, Hamish Campbell, Agustina Gomez-Laich, Patricia Urge, J Walker, Yuzhi Cai, Lianli Gao, Flavio Quintana, Rory Wilson

    Research output: Contribution to journalArticleResearchpeer-review

    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.
    Original languageEnglish
    Pages (from-to)1-7
    Number of pages7
    JournalPLoS One
    Volume9
    Issue number2
    DOIs
    Publication statusPublished - 21 Feb 2014

    Fingerprint

    Love
    Accelerometers
    Animals
    taxonomy
    artificial intelligence
    Learning systems
    animals
    researchers
    Learning algorithms
    Arents
    Research Personnel
    Biotelemetry
    Statistics
    statistics
    methodology
    biotelemetry
    Telemetering
    Ecology
    Telemetry
    telemetry

    Cite this

    Bidder, Owen R ; Campbell, Hamish ; Gomez-Laich, Agustina ; Urge, Patricia ; Walker, J ; Cai, Yuzhi ; Gao, Lianli ; Quintana, Flavio ; Wilson, Rory. / Love thy neighbour : Automatic animal behavioural classification of acceleration data using the k-nearest neighbour algorithm. In: PLoS One. 2014 ; Vol. 9, No. 2. pp. 1-7.
    @article{b68652b93a884b3e83e6e4b13c960dee,
    title = "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 decisionmechanisms 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.",
    author = "Bidder, {Owen R} and Hamish Campbell and Agustina Gomez-Laich and Patricia Urge and J Walker and Yuzhi Cai and Lianli Gao and Flavio Quintana and Rory Wilson",
    year = "2014",
    month = "2",
    day = "21",
    doi = "10.1371/journal.pone.0088609",
    language = "English",
    volume = "9",
    pages = "1--7",
    journal = "PLoS One",
    issn = "1932-6203",
    publisher = "Public Library of Science (PLoS)",
    number = "2",

    }

    Love thy neighbour : Automatic animal behavioural classification of acceleration data using the k-nearest neighbour algorithm. / Bidder, Owen R; Campbell, Hamish; Gomez-Laich, Agustina; Urge, Patricia; Walker, J; Cai, Yuzhi; Gao, Lianli; Quintana, Flavio; Wilson, Rory.

    In: PLoS One, Vol. 9, No. 2, 21.02.2014, p. 1-7.

    Research output: Contribution to journalArticleResearchpeer-review

    TY - JOUR

    T1 - Love thy neighbour

    T2 - Automatic animal behavioural classification of acceleration data using the k-nearest neighbour algorithm

    AU - Bidder, Owen R

    AU - Campbell, Hamish

    AU - Gomez-Laich, Agustina

    AU - Urge, Patricia

    AU - Walker, J

    AU - Cai, Yuzhi

    AU - Gao, Lianli

    AU - Quintana, Flavio

    AU - Wilson, Rory

    PY - 2014/2/21

    Y1 - 2014/2/21

    N2 - 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 decisionmechanisms 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.

    AB - 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 decisionmechanisms 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.

    U2 - 10.1371/journal.pone.0088609

    DO - 10.1371/journal.pone.0088609

    M3 - Article

    VL - 9

    SP - 1

    EP - 7

    JO - PLoS One

    JF - PLoS One

    SN - 1932-6203

    IS - 2

    ER -