Seek and learn: automated identification of microevents in animal behaviour using envelopes of acceleration data and machine learning

1. Animal-borne accelerometers have been used across more than 120 species
to infer biologically significant information such as energy expenditure
and broad behavioural categories. While the accelerometer’s high
sensitivity to movement and fast response times present the unprecedented
opportunity to resolve fine-scale behaviour, leveraging this opportunity
will require overcoming the challenge of developing general, automated
methods to analyse the nonstationary signals generated by nonlinear
processes governing erratic, impulsive movement characteristic of
fine-scale behaviour. 2. We address this issue by conceptualising
fine-scale behaviour in terms of characteristic microevents: impulsive
movements producing brief (<1 s) shock signals in accelerometer
data. We propose a ‘seek-and-learn’ approach: a novel microevent detection
step first locates where shock signals occur (‘seek’) by searching for
peaks in envelopes of acceleration data. Robust machine learning (‘learn’)
employing meaningful features then separates microevents. We showcase the
application of our method on tri-axial accelerometer data collected on ten
free-living meerkats (Suricata suricatta) for four fine-scale foraging
behaviours – searching for digging sites, one-armed digging, two-armed
digging, and head jerks during prey ingestion. Annotated videos served as
groundtruth, and performance was benchmarked against that of a variety of
classical machine-learning approaches. 3. Microevent identification
(μEvId) with eight features in a three-node hierarchical classification
scheme employing logistic regression at each node achieved a mean overall
accuracy of >85% during leave-one-individual-out cross-validation,
and exceeded that of the best classical machine learning approach by 9%.
μEvId was found to be robust not only to inter-individual variation but
also to large changes in model parameters. 4. Our results show that
microevents can be modelled as impulse responses of the animal
body-and-sensor system. The microevent detection step retains only
informative regions of the signal, which results in the selection of
discriminative features that reflect biomechanical differences between
microevents. Moving-window-based classical machine learning approaches
lack this prefiltering step, and were found to be suboptimal for capturing
the nonstationary dynamics of the recorded signals. The general, automated
technique of μEvId, together with existing models that can identify broad
behavioural categories, provides future studies with a powerful toolkit to
exploit the full potential of accelerometers for animal behaviour
recognition.