Development and application of a machine learning algorithm for classification of elasmobranch behaviour from accelerometry data

Abiotic drivers of the space use and activity of gray reef sharks Carcharhinus amblyrhynchos in a dynamic tidal environment

Predators display rhythms in behavior and habitat use, often with the goal of maximizing foraging success. The underlying mechanisms behind these rhythms are generally linked to abiotic conditions related to diel, lunar, or seasonal cycles. To understand their effects on the space use, activity, and swimming depth of gray reef sharks (Carcharhinus amblyrhynchos), we tagged 38 individuals with depth and accelerometer sensors in a French Polynesian atoll channel exposed to strong tidal flow, and monitored them over a year. C. amblyrhynchos used a larger space during nighttime and were more active at night and during outgoing currents. Shark activity also peaked during the full and new moons. The swimming depth of sharks was mostly influenced by diel cycles, with sharks swimming deeper during the day compared to nighttime. The dynamic energyscape may promote the emergence of discrete behavioral strategies in reef sharks that use the south channel of Fakarava for resting and foraging purposes. Turbulence imposed by outgoing tides induces additional foraging cost on sharks, shifting their hunting areas to the southern part of the channel, where turbulence is less pronounced. Understanding when and where sharks are active and foraging is important for our understanding of predator-prey dynamics and ecosystem dynamics. This study highlights how abiotic rhythms in a highly dynamic environment likely generate spatiotemporal heterogeneity in the distribution of predation pressure.

V, Vijayan NT, Subramanian M, Gauns MU. Applications of Machine Learning in Chemical and Biological Oceanography

Machine learning (ML) refers to computer algorithms that predict a meaningful output or categorize complex systems based on a large amount of data. ML is applied in various areas including natural science, engineering, space exploration, and even gaming development. This review focuses on the use of machine learning in the field of chemical and biological oceanography. In the prediction of global fixed nitrogen levels, partial carbon dioxide pressure, and other chemical properties, the application of ML is a promising tool. Machine learning is also utilized in the field of biological oceanography to detect planktonic forms from various images (i.e., microscopy, FlowCAM, and video recorders), spectrometers, and other signal processing techniques. Moreover, ML successfully classified the mammals using their acoustics, detecting endangered mammalian and fish species in a specific environment. Most importantly, using environmental data, the ML proved to be an effective method for predicting hypoxic conditions and harmful algal bloom events, an essential measurement in terms of environmental monitoring. Furthermore, machine learning was used to construct a number of databases for various species that will be useful to other researchers, and the creation of new algorithms will help the marine research community better comprehend the chemistry and biology of the ocean.

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

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.

Short-term behavioural responses of Atlantic bluefin tuna to catch-and-release fishing

Catch-and-release (C&R) angling is often touted as a sustainable form of ecotourism, yet the fine-scale behaviour and physiological responses of released fish is often unknown, especially for hard-to-study large pelagic species like Atlantic bluefin tuna (ABFT; Thunnus thunnus). Multi-channel sensors were deployed and recovered from 10 ABFTs in a simulated recreational C&R event off the west coast of Ireland. Data were recorded from 6 to 25 hours, with one ABFT (tuna X) potentially suffering mortality minutes after release. Almost all ABFTs (n = 9, including tuna X) immediately and rapidly (vertical speeds of ~2.0 m s^-1) made powered descents and used 50-60% of the available water column within 20 seconds, before commencing near-horizontal swimming ~60 seconds post-release. Dominant tailbeat frequency was ~50% higher in the initial hours post-release and appeared to stabilize at 0.8-1.0 Hz some 5-10 hours post-release. Results also suggest different short-term behavioural responses to noteworthy variations in capture and handling procedures (injury and reduced air exposure events). Our results highlight both the immediate and longer-term effects of C&R on ABFTs and that small variations in C&R protocols can influence physiological and behavioural responses of species like the commercially valuable and historically over-exploited ABFT.

Using Drones to Assess Volitional Swimming Kinematics of Manta Ray Behaviors in the Wild

Drones have become increasingly popular tools to study marine megafauna but are underutilized in batoid research. We used drones to collect video data of manta ray (Mobula cf. birostris) swimming and assessed behavior-specific kinematics in Kinovea, a semi-automated point-tracking software. We describe a ‘resting’ behavior of mantas making use of strong currents in man-made inlets in addition to known ‘traveling’ and ‘feeding’ behaviors. No significant differences were found between the swimming speed of traveling and feeding behaviors, although feeding mantas had a significantly higher wingbeat frequency than traveling mantas. Resting mantas swam at a significantly slower speed and wingbeat frequency, suggesting that they were continuously swimming with the minimum effort required to maintain position and buoyancy. Swimming speed and wingbeat frequency of traveling and feeding behaviors overlapped, which could point to other factors such as prey availability and a transitional behavior, influencing how manta rays swim. These baseline swimming kinematic data have valuable applications to other emerging technologies in manta ray research.

Forensic odontology: Assessing bite wounds to determine the role of teeth in piscivorous fishes

Teeth facilitate the acquisition and processing of food in most vertebrates. However, relatively little is known about the functions of the diverse tooth morphologies observed in fishes. Piscivorous fishes (fish-eating fish) are crucial in shaping community structure and rely on their oral teeth to capture and/or process prey. However, how teeth are utilized in capturing and/or processing prey remains unclear. Most studies have determined the function of teeth by assessing morphological traits. The behavior during feeding, however, is seldom quantified. Here, we describe the function of teeth within piscivorous fishes by considering how morphological and behavioral traits interact during prey capture and processing. This was achieved through aquarium-based performance experiments, where prey fish were fed to 12 species of piscivorous fishes. Building on techniques in forensic odontology, we incorporate a novel approach to quantify and categorize bite damage on prey fish that were extracted from the piscivore’s stomachs immediately after being ingested. We then assess the significance of morphological and behavioral traits in determining the extent and severity of damage inflicted on prey fish. Results show that engulfing piscivores capture their prey whole and head-first. Grabbing piscivores capture prey tail-first using their teeth, process them using multiple headshakes and bites, before spitting them out, and then re-capturing prey head-first for ingestion. Prey from engulfers sustained minimal damage, whereas prey from grabbers sustained significant damage to the epaxial musculature. Within grabbers, headshakes were significantly associated with more severe damage categories. Headshaking behavior damages the locomotive muscles of prey, presumably to prevent escape. Compared to non-pharyngognaths, pharyngognath piscivores inflict significantly greater damage to prey. Overall, when present, oral jaw teeth appear to be crucial for both prey capture and processing (immobilization) in piscivorous fishes.

Machine learning accurately predicts the multivariate performance phenotype from morphology in lizards

Completing the genotype-to-phenotype map requires rigorous measurement of the entire multivariate organismal phenotype. However, phenotyping on a large scale is not feasible for many kinds of traits, resulting in missing data that can also cause problems for comparative analyses and the assessment of evolutionary trends across species. Measuring the multivariate performance phenotype is especially logistically challenging, and our ability to predict several performance traits from a given morphology is consequently poor. We developed a machine learning model to accurately estimate multivariate performance data from morphology alone by training it on a dataset containing performance and morphology data from 68 lizard species. Our final, stacked model predicts missing performance data accurately at the level of the individual from simple morphological measures. This model performed exceptionally well, even for performance traits that were missing values for >90% of the sampled individuals. Furthermore, incorporating phylogeny did not improve model fit, indicating that the phenotypic data alone preserved sufficient information to predict the performance based on morphological information. This approach can both significantly increase our understanding of performance evolution and act as a bridge to incorporate performance into future work on phenomics.

Incorporating thermodynamics in predator-prey games predicts the diel foraging patterns of poikilothermic predators

Models of foraging behaviour typically assume that prey do not adapt to temporal variation in predation risk, such as by avoiding foraging at certain times of the day. When this behavioural plasticity is considered-such as in predator-prey games-the role of abiotic factors is usually ignored. An abiotic factor that exerts strong influence on the physiology and behaviour of many animals is ambient temperature, although it is often ignored from game models as it is implicitly assumed that both predators and prey are homothermic. However, poikilotherms' performance may be reduced in cold conditions due to reduced muscle function, limiting the prey-capture ability of predators and the predator-avoidance and foraging abilities of prey. Here, we use a game-theoretic predator-prey model in which diel temperature changes influence foraging gains and costs to predict the evolutionarily stable diel activity of predators. Our model predicts the range of patterns observed in nature, including nocturnal, diurnal, crepuscular and a previously unexplained post-sunset crepuscular pattern observed in some sharks. In general, smaller predators are predicted to be more diurnal than larger ones. The safety of prey when not foraging is critical, explaining why predators in coral reef systems (with safe refuges) may often have different foraging patterns to pelagic predators. We make a range of testable predictions that will enable the further evaluation of this theoretical framework for understanding diel foraging patterns in poikilotherms.

Classifying Goliath Grouper (Epinephelus itajara) Behaviors from a Novel, Multi-Sensor Tag

Inertial measurement unit sensors (IMU; i.e., accelerometer, gyroscope and magnetometer combinations) are frequently fitted to animals to better understand their activity patterns and energy expenditure. Capable of recording hundreds of data points a second, these sensors can quickly produce large datasets that require methods to automate behavioral classification. Here, we describe behaviors derived from a custom-built multi-sensor bio-logging tag attached to Atlantic Goliath grouper (Epinephelus itajara) within a simulated ecosystem. We then compared the performance of two commonly applied machine learning approaches (random forest and support vector machine) to a deep learning approach (convolutional neural network, or CNN) for classifying IMU data from this tag. CNNs are frequently used to recognize activities from IMU data obtained from humans but are less commonly considered for other animals. Thirteen behavioral classes were identified during ethogram development, nine of which were classified. For the conventional machine learning approaches, 187 summary statistics were extracted from the data, including time and frequency domain features. The CNN was fed absolute values obtained from fast Fourier transformations of the raw tri-axial accelerometer, gyroscope and magnetometer channels, with a frequency resolution of 512 data points. Five metrics were used to assess classifier performance; the deep learning approach performed better across all metrics (Sensitivity = 0.962; Specificity = 0.996; F₁-score = 0.962; Matthew’s Correlation Coefficient = 0.959; Cohen’s Kappa = 0.833) than both conventional machine learning approaches. Generally, the random forest performed better than the support vector machine. In some instances, a conventional learning approach yielded a higher performance metric for particular classes (e.g., the random forest had a F₁-score of 0.971 for backward swimming compared to 0.955 for the CNN). Deep learning approaches could potentially improve behavioral classification from IMU data, beyond that obtained from conventional machine learning methods.

Using Machine Learning for Remote Behaviour Classification-Verifying Acceleration Data to Infer Feeding Events in Free-Ranging Cheetahs

Behavioural studies of elusive wildlife species are challenging but important when they are threatened and involved in human-wildlife conflicts. Accelerometers (ACCs) and supervised machine learning algorithms (MLAs) are valuable tools to remotely determine behaviours. Here we used five captive cheetahs in Namibia to test the applicability of ACC data in identifying six behaviours by using six MLAs on data we ground-truthed by direct observations. We included two ensemble learning approaches and a probability threshold to improve prediction accuracy. We used the model to then identify the behaviours in four free-ranging cheetah males. Feeding behaviours identified by the model and matched with corresponding GPS clusters were verified with previously identified kill sites in the field. The MLAs and the two ensemble learning approaches in the captive cheetahs achieved precision (recall) ranging from 80.1% to 100.0% (87.3% to 99.2%) for resting, walking and trotting/running behaviour, from 74.4% to 81.6% (54.8% and 82.4%) for feeding behaviour and from 0.0% to 97.1% (0.0% and 56.2%) for drinking and grooming behaviour. The model application to the ACC data of the free-ranging cheetahs successfully identified all nine kill sites and 17 of the 18 feeding events of the two brother groups. We demonstrated that our behavioural model reliably detects feeding events of free-ranging cheetahs. This has useful applications for the determination of cheetah kill sites and helping to mitigate human-cheetah conflicts.

Disentangling data dependency using cross-validation strategies to evaluate prediction quality of cattle grazing activities using machine learning algorithms and wearable sensor data

Wearable sensors have been explored as an alternative for real-time monitoring of cattle feeding behavior in grazing systems. To evaluate the performance of predictive models such as machine learning (ML) techniques, data cross-validation (CV) approaches are often employed. However, due to data dependencies and confounding effects, poorly performed validation strategies may significantly inflate the prediction quality. In this context, our objective was to evaluate the effect of different CV strategies on the prediction of grazing activities in cattle using wearable sensor (accelerometer) data and ML algorithms. Six Nellore bulls (average live weight of 345 ± 21 kg) had their behavior visually classified as grazing or not-grazing for a period of 15 days. Elastic Net Generalized Linear Model (GLM), Random Forest (RF), and Artificial Neural Network (ANN) were employed to predict grazing activity (grazing or not-grazing) using 3-axis accelerometer data. For each analytical method, three CV strategies were evaluated: holdout, leave-one-animal-out (LOAO), and leave-one-day-out (LODO). Algorithms were trained using similar dataset sizes (holdout: n = 57,862; LOAO: n = 56,786; LODO: n = 56,672). Overall, GLM delivered the worst prediction accuracy (53%) compared to the ML techniques (65% for both RF and ANN), and ANN performed slightly better than RF for LOAO (73%) and LODO (64%) across CV strategies. The holdout yielded the highest nominal accuracy values for all three ML approaches (GLM: 59%, RF: 76%, and ANN: 74%), followed by LODO (GLM: 49%, RF: 61%, and ANN: 63%) and LOAO (GLM: 52%, RF: 57%, and ANN: 57%). With a larger dataset (i.e., more animals and grazing management scenarios), it is expected that accuracy could be increased. Most importantly, the greater prediction accuracy observed for holdout CV may simply indicate a lack of data independence and the presence of carry-over effects from animals and grazing management. Our results suggest that generalizing predictive models to unknown (not used for training) animals or grazing management may incur poor prediction quality. The results highlight the need for using management knowledge to define the validation strategy that is closer to the real-life situation, i.e., the intended application of the predictive model.

Mini-COVIDNet: Efficient Lightweight Deep Neural Network for Ultrasound Based Point-of-Care Detection of COVID-19

Lung ultrasound (US) imaging has the potential to be an effective point-of-care test for detection of COVID-19, due to its ease of operation with minimal personal protection equipment along with easy disinfection. The current state-of-the-art deep learning models for detection of COVID-19 are heavy models that may not be easy to deploy in commonly utilized mobile platforms in point-of-care testing. In this work, we develop a lightweight mobile friendly efficient deep learning model for detection of COVID-19 using lung US images. Three different classes including COVID-19, pneumonia, and healthy were included in this task. The developed network, named as Mini-COVIDNet, was bench-marked with other lightweight neural network models along with state-of-the-art heavy model. It was shown that the proposed network can achieve the highest accuracy of 83.2% and requires a training time of only 24 min. The proposed Mini-COVIDNet has 4.39 times less number of parameters in the network compared to its next best performing network and requires a memory of only 51.29 MB, making the point-of-care detection of COVID-19 using lung US imaging plausible on a mobile platform. Deployment of these lightweight networks on embedded platforms shows that the proposed Mini-COVIDNet is highly versatile and provides optimal performance in terms of being accurate as well as having latency in the same order as other lightweight networks. The developed lightweight models are available at https://github.com/navchetan-awasthi/Mini-COVIDNet.

Using tri-axial accelerometer loggers to identify spawning behaviours of large pelagic fish

BACKGROUND

Tri-axial accelerometers have been used to remotely describe and identify in situ behaviours of a range of animals without requiring direct observations. Datasets collected from these accelerometers (i.e. acceleration, body position) are often large, requiring development of semi-automated analyses to classify behaviours. Marine fishes exhibit many "burst" behaviours with high amplitude accelerations that are difficult to interpret and differentiate. This has constrained the development of accurate automated techniques to identify different "burst" behaviours occurring naturally, where direct observations are not possible.

METHODS

We trained a random forest machine learning algorithm based on 624 h of accelerometer data from six captive yellowtail kingfish during spawning periods. We identified five distinct behaviours (swim, feed, chafe, escape, and courtship), which were used to train the model based on 58 predictive variables.

RESULTS

Overall accuracy of the model was 94%. Classification of each behavioural class was variable; F₁ scores ranged from 0.48 (chafe) - 0.99 (swim). The model was subsequently applied to accelerometer data from eight free-ranging kingfish, and all behaviour classes described from captive fish were predicted by the model to occur, including 19 events of courtship behaviours ranging from 3 s to 108 min in duration.

CONCLUSION

Our findings provide a novel approach of applying a supervised machine learning model on free-ranging animals, which has previously been predominantly constrained to direct observations of behaviours and not predicted from an unseen dataset. Additionally, our findings identify typically ambiguous spawning and courtship behaviours of a large pelagic fish as they naturally occur.

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

Averting today's loss of biodiversity and ecosystem services can be achieved through conservation efforts, especially of keystone species. Giraffes (Giraffa camelopardalis) play an important role in sustaining Africa's ecosystems, but are 'vulnerable' according to the IUCN Red List since 2016. Monitoring an animal's behavior in the wild helps to develop and assess their conservation management. One mechanism for remote tracking of wildlife behavior is to attach accelerometers to animals to record their body movement. We tested two different commercially available high-resolution accelerometers, e-obs and Africa Wildlife Tracking (AWT), attached to the top of the heads of three captive giraffes and analyzed the accuracy of automatic behavior classifications, focused on the Random Forests algorithm. For both accelerometers, behaviors of lower variety in head and neck movements could be better predicted (i.e., feeding above eye level, mean prediction accuracy e-obs/AWT: 97.6%/99.7%; drinking: 96.7%/97.0%) than those with a higher variety of body postures (such as standing: 90.7-91.0%/75.2-76.7%; rumination: 89.6-91.6%/53.5-86.5%). Nonetheless both devices come with limitations and especially the AWT needs technological adaptations before applying it on animals in the wild. Nevertheless, looking at the prediction results, both are promising accelerometers for behavioral classification of giraffes. Therefore, these devices when applied to free-ranging animals, in combination with GPS tracking, can contribute greatly to the conservation of giraffes.

The use of machine learning to detect foraging behaviour in whale sharks: a new tool in conservation

In this study we present the first attempt at modelling the feeding behaviour of whale sharks using a machine learning analytical method. A total of eight sharks were monitored with tri-axial accelerometers and their foraging behaviours were visually observed. Our results highlight that the random forest model is a valid and robust approach to predict the feeding behaviour of the whale shark. In conclusion this novel approach exposes the practicality of this method to serve as a conservation tool and the capability it offers in monitoring potential disturbances of the species.

Application of machine learning to identify predators of stocked fish in Lake Ontario: using acoustic telemetry predation tags to inform management

Understanding predator-prey interactions and food web dynamics is important for ecosystem-based management in aquatic environments, as they experience increasing rates of human-induced changes, such as the addition and removal of fishes. To quantify the post-stocking survival and predation of a prey fish in Lake Ontario, 48 bloater Coregonus hoyi were tagged with acoustic telemetry predation tags and were tracked on an array of 105 acoustic receivers from November 2018 to June 2019. Putative predators of tagged bloater were identified by comparing movement patterns of six species of salmonids (i.e., predators) in Lake Ontario with the post-predated movements of bloater (i.e., prey) using a random forests algorithm, a type of supervised machine learning. A total of 25 bloater (53% of all detected) were consumed by predators on average (± S.D.) 3.1 ± 2.1 days after release. Post-predation detections of predators occurred for an average (± S.D.) of 78.9 ± 76.9 days, providing sufficient detection data to classify movement patterns. Tagged lake trout Salvelinus namaycush provided the most reliable classification from behavioural predictor variables (89% success rate) and was identified as the main consumer of bloater (consumed 50%). Movement networks between predicted and tagged lake trout were significantly correlated over a 6 month period, supporting the classification of lake trout as a common bloater predator. This study demonstrated the ability of supervised learning techniques to provide greater insight into the fate of stocked fishes and predator-prey dynamics, and this technique is widely applicable to inform future stocking and other management efforts.

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

Movement ecology has traditionally focused on the movements of animals over large time scales, but, with advancements in sensor technology, the focus can become increasingly fine scale. Accelerometers are commonly applied to quantify animal behaviours and can elucidate fine-scale (<2 s) behaviours. Machine learning methods are commonly applied to animal accelerometry data; however, they require the trial of multiple methods to find an ideal solution. We used tri-axial accelerometers (10 Hz) to quantify four behaviours in Port Jackson sharks (Heterodontus portusjacksoni): two fine-scale behaviours (<2 s)-(1) vertical swimming and (2) chewing as proxy for foraging, and two broad-scale behaviours (>2 s-mins)-(3) resting and (4) swimming. We used validated data to calculate 66 summary statistics from tri-axial accelerometry and assessed the most important features that allowed for differentiation between the behaviours. One and two second epoch testing sets were created consisting of 10 and 20 samples from each behaviour event, respectively. We developed eight machine learning models to assess their overall accuracy and behaviour-specific accuracy (one classification tree, five ensemble learners and two neural networks). The support vector machine model classified the four behaviours better when using the longer 2 s time epoch (F-measure 89%; macro-averaged F-measure: 90%). Here, we show that this support vector machine (SVM) model can reliably classify both fine- and broad-scale behaviours in Port Jackson sharks.

Early Life Experience Shapes Male Behavior and Social Networks in Drosophila

Living in a group creates a complex and dynamic environment in which behavior of individuals is influenced by and affects the behavior of others. Although social interaction and group living are fundamental adaptations exhibited by many organisms, little is known about how prior social experience, internal states, and group composition shape behavior in groups. Here, we present an analytical framework for studying the interplay between social experience and group interaction in Drosophila melanogaster. We simplified the complexity of interactions in a group using a series of experiments in which we controlled the social experience and motivational states of individuals to compare behavioral patterns and social networks of groups under different conditions. We show that social enrichment promotes the formation of distinct group structure that is characterized by high network modularity, high inter-individual and inter-group variance, high inter-individual coordination, and stable social clusters. Using environmental and genetic manipulations, we show that visual cues and cVA-sensing neurons are necessary for the expression of social interaction and network structure in groups. Finally, we explored the formation of group behavior and structure in heterogenous groups composed of flies with distinct internal states and documented emergent structures that are beyond the sum of the individuals that constitute it. Our results demonstrate that fruit flies exhibit complex and dynamic social structures that are modulated by the experience and composition of different individuals within the group. This paves the path for using simple model organisms to dissect the neurobiology of behavior in complex social environments.

Hybrid Physical Education Teaching and Curriculum Design Based on a Voice Interactive Artificial Intelligence Educational Robot

In order to promote the development of individualized, accurate and intelligent physical education teaching, combined with artificial intelligence technology, the current physical education teaching mode has been improved. Through the establishment of an artificial intelligence educational robot based on voice interaction, a hybrid physical education teaching mode is constructed to realize personalized education for students. First, the speech recognition system is designed from three aspects of speech recognition, interaction management and speech synthesis, and the accuracy of recognition is improved by algorithm. Second, a new mode of hybrid physical education teaching is constructed. Through intelligent information technology, the advantages of traditional physical education teaching are combined to improve the classroom efficiency of physical education teaching and personalized education ability for students. Finally, the relevant experimental scheme and questionnaire are designed, and the actual situation of an educational robot introduced into physical education teaching is investigated and evaluated. The results show that the recognition accuracy of the artificial intelligence speech recognition system can reach more than 90%. It can communicate well with students and answer students’ questions. An educational robot is introduced into physical education teaching, and students’ learning attitude and interest are evaluated. The results show that before and after the introduction of an educational robot in physical education teaching, the average score of students’ learning interest increases by 21 points, and the average score of learning attitude increases by 9.8 points. Therefore, the introduction of an artificial intelligence educational robot based on voice interaction in physical education teaching can help to improve the classroom efficiency of physical education teaching and students’ interest. This study provides a reference for the development of artificial intelligence teaching and promoting the development of artificial intelligence.

Challenges of machine learning model validation using correlated behaviour data: Evaluation of cross-validation strategies and accuracy measures

Automated monitoring of the movements and behaviour of animals is a valuable research tool. Recently, machine learning tools were applied to many species to classify units of behaviour. For the monitoring of wild species, collecting enough data for training models might be problematic, thus we examine how machine learning models trained on one species can be applied to another closely related species with similar behavioural conformation. We contrast two ways to calculate accuracies, termed here as overall and threshold accuracy, because the field has yet to define solid standards for reporting and measuring classification performances. We measure 21 dogs and 7 wolves, and find that overall accuracies are between 51 and 60% for classifying 8 behaviours (lay, sit, stand, walk, trot, run, eat, drink) when training and testing data are from the same species and between 41 and 51% when training and testing is cross-species. We show that using data from dogs to predict the behaviour of wolves is feasible. We also show that optimising the model for overall accuracy leads to similar overall and threshold accuracies, while optimizing for threshold accuracy leads to threshold accuracies well above 80%, but yielding very low overall accuracies, often below the chance level. Moreover, we show that the most common method for dividing the data between training and testing data (random selection of test data) overestimates the accuracy of models when applied to data of new specimens. Consequently, we argue that for the most common goals of animal behaviour recognition overall accuracy should be the preferred metric. Considering, that often the goal is to collect movement data without other methods of observation, we argue that training data and testing data should be divided by individual and not randomly.

Behavioural inference from signal processing using animal-borne multi-sensor loggers: a novel solution to extend the knowledge of sea turtle ecology

The identification of sea turtle behaviours is a prerequisite to predicting the activities and time-budget of these animals in their natural habitat over the long term. However, this is hampered by a lack of reliable methods that enable the detection and monitoring of certain key behaviours such as feeding. This study proposes a combined approach that automatically identifies the different behaviours of free-ranging sea turtles through the use of animal-borne multi-sensor recorders (accelerometer, gyroscope and time-depth recorder), validated by animal-borne video-recorder data. We show here that the combination of supervised learning algorithms and multi-signal analysis tools can provide accurate inferences of the behaviours expressed, including feeding and scratching behaviours that are of crucial ecological interest for sea turtles. Our procedure uses multi-sensor miniaturized loggers that can be deployed on free-ranging animals with minimal disturbance. It provides an easily adaptable and replicable approach for the long-term automatic identification of the different activities and determination of time-budgets in sea turtles. This approach should also be applicable to a broad range of other species and could significantly contribute to the conservation of endangered species by providing detailed knowledge of key animal activities such as feeding, travelling and resting.

Using Artificial Intelligence to Predict Survivability Likelihood and Need for Surgery in Horses Presented With Acute Abdomen (Colic)

Artificial intelligence and machine learning have promising applications in several medical fields of diagnosis, imaging, and laboratory testing procedures. However, the use of this technology in the veterinary medicine field is lagging behind, and there are many areas where it could be used with potentially successful outcomes and results. In this study, two critical predictions were explored in horses presented with acute abdomen (colic) using this technology. Those were the need for surgical intervention and survivability likelihood of affected horses based on clinical data (history, clinical examination findings, and diagnostic procedures). The two prediction parameters were explored using the application of Decision Trees, Multilayer Perceptron, Bayes Network, and Naïve Bayes. The machine learning algorithms were able to predict the need for surgery and survivability likelihood of horses presented with acute abdomen (colic) with 76% and 85% accuracy, respectively. The application of this technology in the different clinical fields of veterinary medicine appears to be of a value and warrants further investigation and testing.

Thermal performance responses in free-ranging elasmobranchs depend on habitat use and body size

Temperature is one of the most influential drivers of physiological performance and behaviour in ectotherms, determining how these animals relate to their ecosystems and their ability to succeed in particular habitats. Here, we analysed the largest set of acceleration data compiled to date for elasmobranchs to examine the relationship between volitional activity and temperature in 252 individuals from 8 species. We calculated activation energies for the thermal performance response in each species and estimated optimum temperatures using an Arrhenius breakpoint analysis, subsequently fitting thermal performance curves to the activity data. Juveniles living in confined nursery habitats not only spent substantially more time above their optimum temperature and at the upper limits of their performance breadths compared to larger, less site-restricted animals, but also showed lower activation energies and broader performance curves. Species or life stages occupying confined habitats featured more generalist behavioural responses to temperature change, whereas wider ranging elasmobranchs were characterised by more specialist behavioural responses. The relationships between the estimated performance regimes and environmental temperature limits suggest that animals in confined habitats, including many juvenile elasmobranchs within nursery habitats, are likely to experience a reduction of performance under a warming climate, although their flatter thermal response will likely dampen this impact. The effect of warming on less site-restricted species is difficult to forecast since three of four species studied here did not reach their optimum temperature in the wild, although their specialist performance characteristics may indicate a more rapid decline should optimum temperatures be exceeded.

Augmenting biologging with supervised machine learning to study in situ behavior of the medusa Chrysaora fuscescens

Zooplankton play critical roles in marine ecosystems, yet their fine-scale behavior remains poorly understood because of the difficulty in studying individuals in situ. Here, we combine biologging with supervised machine learning (ML) to propose a pipeline for studying in situ behavior of larger zooplankton such as jellyfish. We deployed the ITAG, a biologging package with high-resolution motion sensors designed for soft-bodied invertebrates, on eight Chrysaora fuscescens in Monterey Bay, using the tether method for retrieval. By analyzing simultaneous video footage of the tagged jellyfish, we developed ML methods to: (1) identify periods of tag data corrupted by the tether method, which may have compromised prior research findings, and (2) classify jellyfish behaviors. Our tools yield characterizations of fine-scale jellyfish activity and orientation over long durations, and we conclude that it is essential to develop behavioral classifiers on in situ rather than laboratory data.

Summary: High-resolution motion sensors paired with supervised machine learning can be used to infer fine-scale in situ behavior of zooplankton over long durations.

Acoustic accelerometry reveals diel activity patterns in premigratory Port Jackson sharks

Distinguishing the factors that influence activity within a species advances understanding of their behavior and ecology. Continuous observation in the marine environment is not feasible but biotelemetry devices provide an opportunity for detailed analysis of movements and activity patterns. This study investigated the detail that calibration of accelerometers measuring root mean square (RMS) acceleration with video footage can add to understanding the activity patterns of male and female Port Jackson sharks (Heterodontus portusjacksoni) in a captive environment. Linear regression was used to relate RMS acceleration output to time-matched behavior captured on video to quantify diel activity patterns. To validate captive data, diel patterns from captive sharks were compared with diel movement data from free-ranging sharks using passive acoustic tracking. The RMS acceleration data showed captive sharks exhibited nocturnal diel patterns peaking during the late evening before midnight and decreasing before sunrise. Correlation analysis revealed that captive animals displayed similar activity patterns to free-ranging sharks. The timing of wild shark departures for migration in the late breeding season corresponded with elevated diel activity at night within the captive individuals, suggesting a form of migratory restlessness in captivity. By directly relating RMS acceleration output to activity level, we show that sex, time of day, and sex-specific seasonal behavior all influenced activity levels. This study contributes to a growing body of evidence that RMS acceleration data are a promising method to determine activity patterns of cryptic marine animals and can provide more detailed information when validated in captivity.

Activity seascapes highlight central place foraging strategies in marine predators that never stop swimming

BACKGROUND

Central place foragers (CPF) rest within a central place, and theory predicts that distance of patches from this central place sets the outer limits of the foraging arena. Many marine ectothermic predators behave like CPF animals, but never stop swimming, suggesting that predators will incur 'travelling' costs while resting. Currently, it is unknown how these CPF predators behave or how modulation of behavior contributes to daily energy budgets. We combine acoustic telemetry, multi-sensor loggers, and hidden Markov models (HMMs) to generate 'activity seascapes', which combine space use with patterns of activity, for reef sharks (blacktip reef and grey reef sharks) at an unfished Pacific atoll.

RESULTS

Sharks of both species occupied a central place during the day within deeper, cooler water where they were less active, and became more active over a larger area at night in shallower water. However, video cameras on two grey reef sharks revealed foraging attempts/success occurring throughout the day, and that multiple sharks were refuging in common areas. A simple bioenergetics model for grey reef sharks predicted that diel changes in energy expenditure are primarily driven by changes in swim speed and not body temperature.

CONCLUSIONS

We provide a new method for simultaneously visualizing diel space use and behavior in marine predators, which does not require the simultaneous measure of both from each animal. We show that blacktip and grey reef sharks behave as CPFs, with diel changes in activity, horizontal and vertical space use. However, aspects of their foraging behavior may differ from other predictions of traditional CPF models. In particular, for species that never stop swimming, patch foraging times may be unrelated to patch travel distance.