Passenger flow prediction has drawn increasing attention in the deep learning research field due to its great importance in traffic management and public safety. The major challenge of this essential task lies in multiple spatiotemporal correlations that exhibit complex non-linear correlations. Although both the spatial and temporal perspectives have been considered in modeling, most existing works have ignored complex temporal correlations or underlying spatial similarity. In this paper, we identify the unique spatiotemporal correlation of urban metro flow, and propose an attention-based deep spatiotemporal network with multi-task learning (ADST-Net) at a citywide level to predict the future flow from historical observations. ADST-Net uses three independent channels with the same structure to model the recent, daily-periodic and weekly-periodic complicated spatiotemporal correlations, respectively. Specifically, each channel uses the framework of residual networks, the rectified block and the multi-scale convolutions to mine spatiotemporal correlations. The residual networks can effectively overcome the gradient vanishing problem. The rectified block adopts an attentional mechanism to automatically reweigh measurements at different time intervals, and the multi-scale convolutions are used to extract explicit spatial relationships. ADST-Net also introduces an external embedding mechanism to extract the influence of external factors on flow prediction, such as weather conditions. Furthermore, we enforce multi-task learning to utilize transition passenger flow volume prediction as an auxiliary task during the training process for generalization. Through this model, we can not only capture the steady trend, but also the sudden changes of passenger flow. Extensive experimental results on two real-world traffic flow datasets demonstrate the obvious improvement and superior performance of our proposed algorithm compared with state-of-the-art baselines.
Traffic accident prediction on road levels and minute levels plays an important role in optimizing public safety and improving traffic infrastructure. However, there are still some challenges in this work. Firstly, the dynamic factors (e.g. traffic flow) affecting traffic accidents make the road network have dynamic spatio‐temporal dependency, which leads to biased prediction results. Secondly, the occurrence of traffic accidents is a small probability event, which brings about zero‐inflation problem. To address aforementioned problems, the authors propose a Multi‐Attention Dynamic Graph Convolution Network with Cost Sensitive Learning approach (MADGCN). Specifically, in the spatial dimension, MADGCN calculates the attention scores of different types of dynamic factors through attention mechanism to simulate the different influence degrees of different factors, and models dynamic inter‐road spatial correlation through Graph Convolution Network (GCN). In the temporal aspect, MADGCN adaptively models the dynamic temporal correlations through self‐attention blocks. In addition, MADGCN improves the loss function based on cost‐sensitive learning strategy to increase the cost of false classification of positive samples, so as to accurately mine sparse positive samples. Experimental results on two real‐world traffic accident datasets demonstrate the superiority of MADGCN. Compared with the existing state‐of‐the‐art methods, F1‐score of MADGCN on two real‐world traffic accident datasets is improved by 11.61% and 9.15%, respectively.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.