Machine learning goes wild: Using data from captive individuals to infer wildlife behaviours

Abstract

1. Remotely tracking distinct behaviours of animals using acceleration data and machine learning has been carried out successfully in several species in captive settings. In order to study the ecology of animals in natural habitats, such behaviour classification models need to be transferred to wild individuals. However, at present, the development of those models usually requires direct observation of the target animals. 2. The goal of this study was to infer the behaviour of wild, free-roaming animals from acceleration data by training behaviour classification models on captive individuals, without the necessity to observe their wild conspecifics. We further sought to develop methods to validate the credibility of the resulting behaviour extrapolations. 3. We trained two machine learning algorithms proposed by the literature, Random Forest (RF) and Support Vector Machine (SVM), on data from captive red foxes (Vulpes vulpes) and later applied them to data from wild foxes. We also tested a new advance for behaviour classification, by applying a moving window to an Artificial Neural Network (ANN). Finally, we investigated four strategies to validate our classification output. 4. While all three machine learning algorithms performed well under training conditions (Kappa values: RF (0.82), SVM (0.78), ANN (0.85)), the established methods, RF and SVM, failed in classifying distinct behaviours when transferred from captive to wild foxes. Behaviour classification with the ANN and a moving window, in contrast, inferred distinct behaviours and showed consistent results for most individuals. 5. Our approach is a substantial improvement over the methods previously proposed in the literature as it generated plausible results for wild fox behaviour. We were able to infer the behaviour of wild animals that have never been observed in the wild and to further illustrate the credibility of the output. This framework is not restricted to foxes but can be applied to infer the behaviour of many other species and thus empowers new advances in behavioural ecology.

Wildlife Monitoring on the Edge: A Performance Evaluation of Embedded Neural Networks on Microcontrollers for Animal Behavior Classification

“…If this approach is to be used for other semi-wild or wild animals, their habits should be studied and those of interest should be recorded and used to train a neural network in a supervised manner. A recent study [ 26 ] proposed a novel approach for using data from captive individuals to infer wildlife behavior, which could be taken as an inspiration for future works.…”

Behaviour Classification on Giraffes (Giraffa camelopardalis) Using Machine Learning Algorithms on Triaxial Acceleration Data of Two Commonly Used GPS Devices and Its Possible Application for Their Management and Conservation

“…These behaviors as well as measurements of mixed behavior bursts might occur in the wild, but they will not be classified because the algorithms were not trained on them. These undetected behaviors could be narrowed down through a probability threshold and classified as “unknown behavior” [ 25 ]. Anyway, testing the unlabeled data of numerous wild giraffes against the patterns of our study’s few individuals surely will lead to a higher percentage of behavior misclassifications.…”

“…This makes it possible that, in principle, models based on few individuals can also provide reliable behavioral predictions. Rast et al [ 25 ] were able to infer the behavior of wild foxes using acceleration data, a supervised machine learning algorithm (ANN), and a training data set obtained from only two captive individuals.…”

“…To train the algorithms for pattern recognition and data classification, the acquisition of acceleration data is coupled with direct observation of the behaviors of the tagged animals. Using one portion of this observation-truthed data set to train the algorithm and another portion to infer behavior from it, allows validation of the predicted behavior [ 25 ]. This technique of behavioral detection by analyzing acceleration data has been successfully used on a variety of bird species [ 23 , 26 , 27 , 28 , 29 , 30 , 31 ], on marine animals [ 32 , 33 , 34 , 35 , 36 , 37 ], and on terrestrial mammals [ 25 , 38 , 39 , 40 , 41 , 42 , 43 , 44 , 45 ].…”

Assessment of Machine Learning Models to Identify Port Jackson Shark Behaviours Using Tri-Axial Accelerometers

“…Additionally, behaviours including gliding, burst swimming, rolling and foraging (pre-prey capture) were recorded for too few instances to train a robust model for tagged individuals. The models developed here considered the behaviours of four individual sharks within a captive setting to allow for the final model to be applied to other individuals; however, it is possible that further individual variation will be present in wild populations and how successful the model is in the wild context may depend on this [ 55 ]. Indeed, the models may require further validation in the wild perhaps in combination with footage from underwater drones.…”