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

Owen R Bidder,Lianli Gao,Yuzhi Cai,James Walker,Agustina Gómez-Laich,Flavio Quintana,Patricia Urgé,Rory P Wilson,Hamish A Campbell,Allan V Kalueff

doi:10.1371/journal.pone.0088609

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.

Highlights

The use of animal attached sensors for monitoring animal movements and behaviour is common practice
Results from the K – Nearest Neighbour (KNN) analysis were compared to the actual behavioural classification of the data in order to obtain overall accuracy
All 5 behaviours were detected using the KNN method trialled on all species, except for the kangaroo which was not tested for the ‘Sit’ behaviour because its incidence was not discernible from the video footage

Summary

Introduction

The use of animal attached sensors for monitoring animal movements and behaviour is common practice (see [1] for review). An increasing number of studies are making use of accelerometers to quantify animal behaviour [14,15,16,17]. Most of these studies identify behaviour following the principles set out in Shepard et al. Body posture can be detected as ‘static’ acceleration, and relates to the orientation of the accelerometer with respect to gravity. Body motion is detected as ‘dynamic’ acceleration when the inertia produced by animal movement registers characteristic signals on the device [18]. Modern accelerometer-equipped data loggers are able to record at rates as high as 300 Hz [19], so manual identification of behavioural patterns in accelerometer data using this approach is arduous and, with increases in the use and capacity of the technology, is set to become more so

Objectives

Methods

Results

Conclusion