Stochastic Origin-Destination Matrix Forecasting Using Dual-Stage Graph Convolutional, Recurrent Neural Networks

Hu, Jilin; Yang, Bin; Guo, Chenjuan; Jensen, Christian S.; Xiong, Hui

doi:10.1109/icde48307.2020.00126

Cited by 72 publications

(38 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Graph convolutional neural networks (GCNNs) have been widely used to predict ETA [11] [16][5] [6], where "Graph" represents the required roadmaps. All of these methods accept spatiotemporal data.…”

Section: Deep Learning Methodsmentioning

confidence: 99%

“…All of these methods accept spatiotemporal data. In addition, a GCNN-based origin-destination method [11] was proposed for ETA estimation. An origin-destination method is not concerned with paths.…”

Section: Deep Learning Methodsmentioning

confidence: 99%

“…Several deep-learning-based methods have been proposed for solving the ETA problem [21] [19] [11] [16][5] [6]. A seminal method is DeepTTE [21], which uses historical trajectories as training data and predicts the ETA of a given trajectory.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

HyperETA: An Estimated Time of Arrival Method based on Hypercube Clustering

Hsu¹

2021

Preprint

View full text Add to dashboard Cite

The Estimated Time of Arrival (ETA) that predict the travel time of a given GPS trajectory has been extensively used in route planning. Deep learning has been widely applied to ETA prediction. However, prediction tasks involve some challenges, such as small data size, low GPU’s precision, high training loss, and low accuracy. Herein, we present a new machine-learning algorithm called HyperETA for ETA prediction. HyperETA is based on a extraordinary clustering method, called Hypercube Clustering. We conducted experiments to compare HyperETA with a deep-learning-based method called DeepTTE by using taxi trajectories as a benchmark. Two variations of both methods were evaluated. The results indicated that HyperETA outperformed the deep-learning approach in terms of prediction accuracy.

show abstract

Section: Deep Learning Methodsmentioning

confidence: 99%

Section: Deep Learning Methodsmentioning

confidence: 99%

See 1 more Smart Citation

HyperETA: An Estimated Time of Arrival Method based on Hypercube Clustering

Hsu¹

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Toqué, Côme, El Mahrsi, and Oukhellou (2016) were among the first to use the long short-term memory (LSTM) neural network model to estimate future OD flows by using the historic OD flows as input. Since their work, more complex deep learning models have been proposed to estimate dynamic future flows, among them convolutional LSTM (Duan et al, 2019), contextualized spatial-temporal networks (Liu et al, 2019), dual-stage graph convolutional recurrent neural networks (Hu, Yang, Guo, Jensen, & Xiong, 2020), spatial-temporal LSTM (Li et al, 2020), spatial-temporal encoder-decoder residual multi-graph convolutional networks (Ke et al, 2021), and dynamic node-edge attention networks (Zhang, Xiao, Shen, & Zhong, 2021). Therefore, dynamic OD networks provide a more comprehensive and detailed day-to-day description for urban dynamics and have become a powerful tool for understanding collective dynamics in recent years.…”

Section: Rel Ated Workmentioning

confidence: 99%

Recurrent origin–destination network for exploration of human periodic collective dynamics

Chen

Xie

Xiao

et al. 2021

Transactions in GIS

View full text Add to dashboard Cite

show abstract

“…The functions of structures can be more easily confirmed through the analysis of interactions of these structures in different periods. In road traffic networks [19][20][21] , scholars could make route recommendations or reachability queries for users by combining the historical data of the networks. In social networks [22][23][24] , scholars could characterize the relationships between users more precisely by recording their specific interactions.…”

mentioning

confidence: 99%

Node Embedding Research Over Temporal Graph

Wu¹,

Wu²,

Yuan³

et al. 2021

International Journal of Software and Informatics

View full text Add to dashboard Cite

Compared with conventional graph data analysis methods, the graph embedding algorithm provides a new graph data analysis strategy. It aims to encode graph nodes into vectors to mine or analyze graph data more effectively using neural networks related technologies. Some classic tasks have been improved significantly by graph embedding methods, such as node classification, link prediction, and traffic flow prediction. Although substantial breakthroughs have been made by former researchers in graph embedding, the nodes embedding problem over temporal graph have been seldom studied. In this work, we propose an Adaptive Temporal Graph EmBebding (ATGEB), attempting to encode temporal graph nodes into vectors by combining previous research and the information propagation characteristics. First, an adaptive cluster method is proposed by solving the situation that nodes active frequency varies types of graph. Then, a new node walk strategy is designed in order to store the time sequence between nodes, and also the walking list will be stored in a bidirectional multi-tree in the walking process to get complete walking lists quickly. Last, based on the basic walking characteristics and graph topology, an important node sampling strategy is proposed to train the satisfied neural network as soon as possible. Sufficient experiments demonstrate that the proposed method surpasses existing embedding methods in terms of node clustering, reachability prediction, and node classification in temporal graphs.

show abstract

Stochastic Origin-Destination Matrix Forecasting Using Dual-Stage Graph Convolutional, Recurrent Neural Networks

Cited by 72 publications

References 31 publications

HyperETA: An Estimated Time of Arrival Method based on Hypercube Clustering

HyperETA: An Estimated Time of Arrival Method based on Hypercube Clustering

Recurrent origin–destination network for exploration of human periodic collective dynamics

Node Embedding Research Over Temporal Graph

Contact Info

Product

Resources

About