Spatio-temporal Awareness Research Articles

LGSA: A next POI prediction method by using local and global interest with spatiotemporal awareness

LGSA: A next POI prediction method by using local and global interest with spatiotemporal awareness

Big data driven vessel trajectory prediction based on sparse multi-graph convolutional hybrid network with spatio-temporal awareness

The revolutionary progress of deep learning and data mining technology has greatly contributed to the rapid development of maritime Internet of Things. In order to advance intelligent maritime traffic services, it is necessary to accurately predict the future trajectories of vessels, which is conducive to collision avoidance, maritime surveillance and abnormal behavior detection. In this work, a new method of vessel trajectory prediction based on sparse multi-graph convolutional hybrid network with spatio-temporal awareness is proposed, namely SMCHN model, which uses sparse spatial graph to capture the adaptive interactions between vessels and remove redundant interactions, and sparse temporal graph can be used to simulate the movement trends of vessels, bringing the movement behavior of vessels closer and closer to reality. In addition, the hybrid network is designed to fuse information from different sources by adjusting weights. Finally, the temporal convolutional network with gating mechanism is used to predict the future trajectories of vessels. Experimental results show that the SMCHN model improves 31%, 66% and 54% in short-term, medium-term and long-term vessel trajectory prediction compared with the optimal baseline model.

Stream Temperature Prediction in a Shifting Environment: Explaining the Influence of Deep Learning Architecture

AbstractStream temperature is a fundamental control on ecosystem health. Recent efforts incorporating process guidance into deep learning models for predicting stream temperature have been shown to outperform existing statistical and physical models. This performance is in part because deep learning architectures can actively learn spatiotemporal relationships that govern how water and energy propagate through a river network. However, exploration of how spatiotemporal awareness and process guidance influence a model's generalizability under shifting environmental conditions such as climate change is limited. Here, we use Explainable Artificial Intelligence (XAI) to interrogate how differing deep learning architectures affect a model's learned spatial and temporal dependencies, and how those learned dependencies affect a model's ability to maintain high accuracy when applied to unseen environmental conditions. Using the Delaware River Basin in the northeastern United States as a test case, we compare two spatiotemporally aware process‐guided deep learning models for predicting stream temperature (a recurrent graph convolution network—RGCN, and a temporal convolution graph model—Graph WaveNet). Both models achieve equally high predictive performance when testing data are well represented in the training data (test root mean squared errors of 1.64°C and 1.65°C); however, Graph WaveNet significantly outperforms RGCN in 4 out of 5 experiments where test partitions represent different types of unseen environmental conditions. XAI results show that the architecture of Graph WaveNet leads to learned spatial relationships with greater fidelity to physical processes, and that this fidelity improves the generalizability of the model when applied to shifting and/or unseen environmental conditions.

Matching user accounts with spatio-temporal awareness across social networks

• A user identification method that considers the spatiotemporal correlation and the inconsistency of check-in records. • A KDE-based method to measure the proximity of two users’ check-in records. • A method to penalize the similarity based on the number of conflictive check-in records. User identification aims at matching user accounts across social sites, which benefits many real-world applications. Existing works based on user trajectories usually address spatial and temporal data separately while not fully utilizing the coupling relation between them. Differently, in this work, we jointly consider spatialtemporal information in users’ acitvities to improve the user identification method. In particular, we observe that check-in records of different users tend to create inconsistent spatialtemporal information. These inconsistencies are useful for eliminating false user matching. Inspired by this observation, we propose a novel user identification method that captures the correlation of spatial and temporal information and the inconsistency in check-in records. It contains three main steps. 1) We measure the similarity of users’ trajectories based on a kernel density estimation, which considers spatial and temporal information simultaneously. 2) We assign a weight to each check-in record to favor discriminative ones. 3) We utilize the inconsistency among check-in records to compute penalties for trajectory similarity. The pair of accounts with higher similarity (than a predefined threshold) is then considered to be from the same user. We evaluate our approach on three ground-truth datasets. The results show that the proposed method offers competitive performance, with F1 values reaching 86.12%, 85.08% and 78.34%, which demonstrates the superiority of the proposed method over state-of-the-art methods.