A comparative analysis of animal behavior (e.g., male vs. female groups) has been widely used to elucidate behavior specific to one group since pre-Darwinian times. However, big data generated by new sensing technologies, e.g., GPS, makes it difficult for them to contrast group differences manually. This study introduces DeepHL, a deep learning-assisted platform for the comparative analysis of animal movement data, i.e., trajectories. This software uses a deep neural network based on an attention mechanism to automatically detect segments in trajectories that are characteristic of one group. It then highlights these segments in visualized trajectories, enabling biologists to focus on these segments, and helps them reveal the underlying meaning of the highlighted segments to facilitate formulating new hypotheses. We tested the platform on a variety of trajectories of worms, insects, mice, bears, and seabirds across a scale from millimeters to hundreds of kilometers, revealing new movement features of these animals.

Many applications use recommender systems to predict user preferences, improve user experience, and increase the amount of sales. However, because of the cold-start problem, it is not easy to recommend items to new users accurately. Recommendation performance degrades in the case of users with little interaction, in particular latent users who have never used the service. To alleviate the cold-start problem, we develop a framework that combines an online shopping domain with information from an Ads platform. Our framework employs deep learning to build a cross-domain recommender system based on shared users in these two domains. We apply Word2Vec to turn textual information on users and items into latent vectors as their representations. The experimental results show the effectiveness of deep neural approaches with knowledge transferred from another domain for the cold-start problem. Textual information may contain useless information, and Word2Vec cannot capture some structural and semantic correlations between different users. Therefore, we propose R-metapath2Vec to enhance user modeling and use the Stacking model to integrate these two kinds of user representations. The experimental results demonstrate the effectiveness of our integration model: our framework can recommend products to users of another domain through Ads distribution in a more accurate level.

Unravelling the secrets of wild animals is one of the biggest challenges in ecology, with bio-logging (i.e., the use of animal-borne loggers or bio-loggers) playing a pivotal role in tackling this challenge. Bio-logging allows us to observe many aspects of animals’ lives, including their behaviours, physiology, social interactions, and external environment. However, bio-loggers have short runtimes when collecting data from resource-intensive (high-cost) sensors. This study proposes using AI on board video-loggers in order to use low-cost sensors (e.g., accelerometers) to automatically detect and record complex target behaviours that are of interest, reserving their devices’ limited resources for just those moments. We demonstrate our method on bio-loggers attached to seabirds including gulls and shearwaters, where it captured target videos with 15 times the precision of a baseline periodic-sampling method. Our work will provide motivation for more widespread adoption of AI in bio-loggers, helping us to shed light onto until now hidden aspects of animals’ lives.

Machine learning‐based behaviour classification using acceleration data is a powerful tool in bio‐logging research. Deep learning architectures such as convolutional neural networks (CNN), long short‐term memory (LSTM) and self‐attention mechanism as well as related training techniques have been extensively studied in human activity recognition. However, they have rarely been used in wild animal studies. The main challenges of acceleration‐based wild animal behaviour classification include data shortages, class imbalance problems, various types of noise in data due to differences in individual behaviour and where the loggers were attached and complexity in data due to complex animal‐specific behaviours, which may have limited the application of deep learning techniques in this area. To overcome these challenges, we explored the effectiveness of techniques for efficient model training: data augmentation, manifold mixup and pre‐training of deep learning models with unlabelled data, using datasets from two species of wild seabirds and state‐of‐the‐art deep learning model architectures. Data augmentation improved the overall model performance when one of the various techniques (none, scaling, jittering, permutation, time‐warping and rotation) was randomly applied to each data during mini‐batch training. Manifold mixup also improved model performance, but not as much as random data augmentation. Pre‐training with unlabelled data did not improve model performance. The state‐of‐the‐art deep learning models, including a model consisting of four CNN layers, an LSTM layer and a multi‐head attention layer, as well as its modified version with shortcut connection, showed better performance among other comparative models. Using only raw acceleration data as inputs, these models outperformed classic machine learning approaches that used 119 handcrafted features. Our experiments showed that deep learning techniques are promising for acceleration‐based behaviour classification of wild animals and highlighted some challenges (e.g. effective use of unlabelled data). There is scope for greater exploration of deep learning techniques in wild animal studies (e.g. advanced data augmentation, multimodal sensor data use, transfer learning and self‐supervised learning). We hope that this study will stimulate the development of deep learning techniques for wild animal behaviour classification using time‐series sensor data.

This paper proposes a new Wi-Fi based indoor positioning method that is robust over unstable Wi-Fi access points (APs). Because Wi-Fi based indoor positioning relies on unstable and uncontrollable infrastructure (Wi-Fi APs), the positioning performance significantly decreases when such unstable APs are included in the localization system. This paper proposes a indoor positioning method by employing ensemble of weak position estimators, which permits us to construct a robust positioning model. Our proposed boosted position estimator has the following features. 1) The estimator does not overfit the training data and thus it is robust over unstable signals from APs. 2) Because each weak estimator employs a small number of APs for positioning, the estimator is not affected by the curse of dimensionality. 3) Our model can adaptively change the weight (importance) of each weak estimator according to a user's position in order to achieve a position-aware precise localization.