STGC-GNNs: A GNN-based traffic prediction framework with a spatial–temporal Granger causality graph

He, Silu; Luo, Qinyao; Du, Ruxu; Zhao, Ling; He, Guangjun; Han, Fei; Li, Haifeng

doi:10.1016/j.physa.2023.128913

Cited by 22 publications

(4 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[36] proposed an estimation method based on a spatiotemporal GAN model, which can predict the impact of planning implementation on urban traffic status given an urban development plan and historical observation data of road networks. The model is based on a conditional GAN model, which takes various travel demands as the input conditions, while modeling the time dependence of traffic flow [37][38][39] at different times of the day using a selfattention mechanism.…”

Section: Gan-based Methodsmentioning

confidence: 99%

Conditional Diffusion Model for Urban Morphology Prediction

Shi,

Zhao,

Liu

et al. 2024

Remote Sensing

Self Cite

View full text Add to dashboard Cite

Predicting urban morphology based on local attributes is an important issue in urban science research. The deep generative models represented by generative adversarial network (GAN) models have achieved impressive results in this area. However, in such methods, the urban morphology is assumed to follow a specific probability distribution and be able to directly approximate the distribution via GAN models, which is not a realistic strategy. As demonstrated by the score-based model, a better strategy is to learn the gradient of the probability distribution and implicitly approximate the distribution. Therefore, in this paper, an urban morphology prediction method based on the conditional diffusion model is proposed. Implementing this approach results in the decomposition of the attribute-based urban morphology prediction task into two subproblems: estimating the gradient of the conditional distribution, and gradient-based sampling. During the training stage, the gradient of the conditional distribution is approximated by using a conditional diffusion model to predict the noise added to the original urban morphology. In the generation stage, the corresponding conditional distribution is parameterized based on the noise predicted by the conditional diffusion model, and the final prediction result is generated through iterative sampling. The experimental results showed that compared with GAN-based methods, our method demonstrated improvements of 5.5%, 5.9%, and 13.2% in the metrics of low-level pixel features, shallow structural features, and deep structural features, respectively.

show abstract

Section: Gan-based Methodsmentioning

confidence: 99%

Conditional Diffusion Model for Urban Morphology Prediction

Shi,

Zhao,

Liu

et al. 2024

Remote Sensing

Self Cite

View full text Add to dashboard Cite

show abstract

“…Due to the irregularity of graph structures, traditional deep learning models are unable to handle this type of data. The emergence of Graph Neural Networks (GNN) [21] has attracted extensive attention from scholars. GNN and Graph Convolutional Network (GCN) [22] demonstrate remarkable capabilities in dealing with unstructured data, making various spatiotemporal learning models based on GNN and GCN become the trend and focus of research in the field of urban road traffic flow and pedestrian flow prediction.…”

Section: ) City Traffic Predictionmentioning

confidence: 99%

“…He et al [21] proposed STGC-GNNs, which can utilize the global-dynamic information for long-term prediction. Zhao et al [22] proposed a combined prediction method called T-GCN, which integrates Graph Convolutional Networks (GCN) and GRU (Gate Recurrent Unit) [23].…”

Section: ) City Traffic Predictionmentioning

confidence: 99%

The Prediction of Flow in Railway Station Based on RRC-STGCN

Wang,

Bai,

Meng

et al. 2023

IEEE Access

View full text Add to dashboard Cite

Prediction of the passenger flow is of great importance to the management and safety in railway stations. Accurate prediction of passenger flow facilitates effective allocation of staff tasks, enhances the efficient utilization of waiting areas, ensures passenger safety, and promotes a smooth travel experience for passengers. However, accurately establishing the spatial-temporal relationship of passenger flow within the station and predicting the passenger flow in each region and time period is challenging due to the varying waiting habits and train preferences of each individual passenger. In this paper, we propose a Residual-RNN-Channel Spatial-Temporal Graph Convolutional Network (RRC-STGCN), which utilizes the channel attention mechanism and residual structure. The model divides the data with a time dimension into multiple periods, capturing spatial-temporal correlations through the channel attention mechanism, and extracting spatial-temporal dependencies from the feature maps using the spatial-temporal convolution module. The model uses a residual structure to fuse features in order to enhance the accuracy of prediction results. In addition, we conduct a comprehensive experimental evaluation using a real dataset of railway station passenger, demonstrating that the RRC-STGCN model outperforms five well-known baselines. Moreover, we provide visualizations of the prediction results, effectively showcasing the dynamic changes in passenger flow in each waiting area.

show abstract

“…DMGCRNs [14] utilize hyperbolic GNNs to capture multi-scale spatial relationships. Different graph structures that can model spatial correlations from different perspectives have also been proposed, including pre-defined spatial graphs, spatial-temporal fusion graphs [15], localized spatial-temporal graphs [16], causality graphs [17], and adaptive multi-level fusion graphs [18].…”

Section: Introductionmentioning

confidence: 99%

Self-Supervised Spatiotemporal Masking Strategy-Based Models for Traffic Flow Forecasting

Liu,

He,

Han

et al. 2023

Symmetry

Self Cite

View full text Add to dashboard Cite

Traffic flow forecasting is an important function of intelligent transportation systems. With the rise of deep learning, building traffic flow prediction models based on deep neural networks has become a current research hotspot. Most of the current traffic flow prediction methods are designed from the perspective of model architectures, using only the traffic features of future moments as supervision signals to guide the models to learn the spatiotemporal dependence in traffic flow. However, traffic flow data themselves contain rich spatiotemporal features, and it is feasible to obtain additional self-supervised signals from the data to assist the model to further explore the underlying spatiotemporal dependence. Therefore, we propose a self-supervised traffic flow prediction method based on a spatiotemporal masking strategy. A framework consisting of symmetric backbone models with asymmetric task heads were applied to learn both prediction and spatiotemporal context features. Specifically, a spatiotemporal context mask reconstruction task was designed to force the model to reconstruct the masked features via spatiotemporal context information, so as to assist the model to better understand the spatiotemporal contextual associations in the data. In order to avoid the model simply making inferences based on the local smoothness in the data without truly learning the spatiotemporal dependence, we performed a temporal shift operation on the features to be reconstructed. The experimental results showed that the model based on the spatiotemporal context masking strategy achieved an average prediction performance improvement of 1.56% and a maximum of 7.72% for longer prediction horizons of more than 30 min compared with the backbone models.

show abstract

STGC-GNNs: A GNN-based traffic prediction framework with a spatial–temporal Granger causality graph

Cited by 22 publications

References 42 publications

Conditional Diffusion Model for Urban Morphology Prediction

Conditional Diffusion Model for Urban Morphology Prediction

The Prediction of Flow in Railway Station Based on RRC-STGCN

Self-Supervised Spatiotemporal Masking Strategy-Based Models for Traffic Flow Forecasting

Contact Info

Product

Resources

About