Background Our understanding of movement patterns and behaviours of wildlife has advanced greatly through the use of improved tracking technologies, including application of accelerometry (ACC) across a wide range of taxa. However, most ACC studies either use intermittent sampling that hinders continuity or continuous data logging relying on tracker retrieval for data downloading which is not applicable for long term study. To allow long-term, fine-scale behavioural research, we evaluated a range of machine learning methods for their suitability for continuous on-board classification of ACC data into behaviour categories prior to data transmission. Methods We tested six supervised machine learning methods, including linear discriminant analysis (LDA), decision tree (DT), support vector machine (SVM), artificial neural network (ANN), random forest (RF) and extreme gradient boosting (XGBoost) to classify behaviour using ACC data from three bird species (white stork Ciconia ciconia, griffon vulture Gyps fulvus and common crane Grus grus) and two mammals (dairy cow Bos taurus and roe deer Capreolus capreolus). Results Using a range of quality criteria, SVM, ANN, RF and XGBoost performed well in determining behaviour from ACC data and their good performance appeared little affected when greatly reducing the number of input features for model training. On-board runtime and storage-requirement tests showed that notably ANN, RF and XGBoost would make suitable on-board classifiers. Conclusions Our identification of using feature reduction in combination with ANN, RF and XGBoost as suitable methods for on-board behavioural classification of continuous ACC data has considerable potential to benefit movement ecology and behavioural research, wildlife conservation and livestock husbandry.

The original random forests algorithm has been widely used and has achieved excellent performance for the classification and regression tasks. However, the research on the theory of random forests lags far behind its applications. In this paper, to narrow the gap between the applications and theory of random forests, we propose a new random forests algorithm, called random Shapley forests (RSFs), based on the Shapley value. The Shapley value is one of the well-known solutions in the cooperative game, which can fairly assess the power of each player in a game. In the construction of RSFs, RSFs uses the Shapley value to evaluate the importance of each feature at each tree node by computing the dependency among the possible feature coalitions. In particular, inspired by the existing consistency theory, we have proved the consistency of the proposed random forests algorithm. Moreover, to verify the effectiveness of the proposed algorithm, experiments on eight UCI benchmark datasets and four real-world datasets have been conducted. The results show that RSFs perform better than or at least comparable with the existing consistent random forests, the original random forests and a classic classifier, support vector machines.

Although tremendous advancement regarding the highly stable metal halide perovskite nanocrystals (PNCs) has been achieved, previous studies were primarily focused on green light-emitting PNCs (i.e., bromine-based PNCs). The stability of chlorine-based PNCs with a violet or blue emission property was still lagging behind that of their bromine-based counterparts. Herein, a nondemanding and versatile strategy for in situ encapsulating allinorganic chlorine-based PNCs with multifold exceptionally high stabilities was presented. Wellordered mesoporous silica enabled the confined growth of PNCs in its pores followed by the porosity sealing by tetramethyl orthosilicate hydrolysis, thereby rendering full encapsulation of chlorine-based PNCs in dense silica that originated from high-temperature calcination. This judiciously designed structure imparted enclosed violet/blue emitting PNCs impart with outstanding long-term stability (>1.5 year) with high photoluminescence quantum yield (i.e., 30.4%) in pure water as the result of complete isolation of PNCs from detrimental stimuli, eventually leading to the application in the white light-emitting diode device.

The continuous development and improvement of tracking devices has enabled researchers to study animal movement ecology, physiology and behaviours in ever-increasing detail (Williams et al., 2019;Wilson et al., 2019). In addition to positional data, various micro-sensors on-board of tracking devices enable the monitoring of different aspects of the wildlife tracked as well as their environment (Ropert-Coudert & Wilson, 2005). Accelerometer (ACC) data is one such feature measured with micro-sensors that is increasingly used to study animal behaviours and energetics (e.g. Williams et al., 2014).

Increasingly animal behaviour studies are enhanced through the use of accelerometry. To allow translation of raw accelerometer data to animal behaviours requires the development of classifiers. Here, we present the “rabc” package to assist researchers with the interactive development of such animal-behaviour classifiers based on datasets consisting out of accelerometer data with their corresponding animal behaviours. Using an accelerometer and a corresponding behavioural dataset collected on white stork (Ciconia ciconia), we illustrate the workflow of this package, including raw data visualization, feature calculation, feature selection, feature visualization, extreme gradient boost model training, validation, and, finally, a demonstration of the behaviour classification results.