Identification of reindeer fine-scale foraging behaviour using tri-axial accelerometer data

Abstract

Animal behavioural responses to the environment ultimately affect their survival. Monitoring animal fine-scale behaviour may improve understanding of animal functional response to the environment and provide an important indicator of the welfare of both wild and domesticated species. In this study, we illustrate the application of collar-attached acceleration sensors for investigating reindeer fine-scale behaviour. Using data from 19 reindeer, we tested the supervised machine learning algorithms Random forests, Support vector machines, and hidden Markov models to classify reindeer behaviour into seven classes: grazing, browsing low from shrubs or browsing high from trees, inactivity, walking, trotting, and other behaviours. We implemented leave-one-subject-out cross-validation to assess generalizable results on new individuals. Our main results illustrated that hidden Markov models were able to classify collar-attached accelerometer data into all our pre-defined behaviours of reindeer with reasonable accuracy while Random forests and Support vector machines were biased towards dominant classes. Random forests using 5-s windows had the highest overall accuracy (85%), while hidden Markov models were able to best predict individual behaviours and handle rare behaviours such as trotting and browsing high. We conclude that hidden Markov models provide a useful tool to remotely monitor reindeer and potentially other large herbivore species behaviour. These methods will allow us to quantify fine-scale behavioural processes in relation to environmental events.

Predictive model of sperm whale prey capture attempts from time-depth data

“…Model predictive performance was evaluated using multiple accuracy measures ( [49,50]; Additional file 1: Table A3): AUC -the area under the ROC (receiver operating characteristic) curve, Sensitivity -the proportion of presences correctly predicted, Specificity -the proportion of absences correctly predicted, and Precision -the proportion of true positives to total predicted positives [51], which were calculated by generating a ROC curve using R package "ROCR" [52]. The AUC index ranges from 0 to 1; AUC ≤ 0.6 indicate a discrimination ability no better than random, 0.6-0.7 indicate moderate predictive performance, 0.7-0.8 as good, 0.8-0.9 as very good and > 0.9 as excellent [53].…”

Classification of behaviors of free-ranging cattle using accelerometry signatures collected by virtual fence collars

“…There are numerous ways to analyze accelerometry data, ranging from simple decision trees to complex neural networks (Riaboff et al, 2022). Most studies utilize supervised machine learning methods such as random forest (de Weerd et al, 2015;Williams et al, 2020;Riaboff et al, 2022), while others use unsupervised machine learning such as hidden Markov Models (Leos- Barajas et al, 2017;Chimienti et al, 2021;Rautiainen et al, 2022). Supervised methods provide the advantage of allowing for accelerometry data to be calibrated on actual behavioral observations, which then allows for prediction of behaviors based on collected data.…”

“…This may result in additional variability in orientation-dependent features, which may render these features unusable without correcting for orientation (Williams et al, 2017;Barker et al, 2018;Kamminga et al, 2018). This sensitivity can be challenging when standardizing accelerometer sensor placement during animal handling, and recapturing individuals to manually fix issues is difficult (Chakravarty et al, 2019;Cade et al, 2021;Rautiainen et al, 2022). To remedy this problem, orientation independent features can be utilized.…”