DyGCN-LSTM: A dynamic GCN-LSTM based encoder-decoder framework for multistep traffic prediction

Kumar, Rahul; Moreira, João Mendes; Chandra, Joydeep

doi:10.1007/s10489-023-04871-3

Cited by 16 publications

(3 citation statements)

References 57 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, different types of adjacency matrices have been proposed to construct multiple graphs for GNNs, such as matrices based on the spatial distance between nodes [33,34]; localized neighborhood connectivity [26,35,36]; similarity [15,33], which is the temporal graph-based correlations between the time series of pairs of nodes; and betweenness [33], which determines the degree of busyness of a road section passed through by all shortest routes between a pair of nodes. We use the dynamic connectivity adjacency matrix A C (t).…”

Section: Traffic Network Graphsmentioning

confidence: 99%

“…Due to its competitive performance, the LSTM method is often regarded as a baseline in subsequently proposed approaches. Meanwhile, ever-more enhanced LSTM models [10,11] and hybrid LSTM-based models [12][13][14][15][16][17] have been proposed.…”

Section: Introductionmentioning

confidence: 99%

“…Graph WaveNet [27] uses a self-adaptive adjacency matrix to capture hidden spatial dependencies as well as dilated casual convolution networks to capture long temporal trends. In [15], DyGCN uses a dynamic distance correlation temporal graphbased adjacency matrix as an alternative to the static spatial distance matrix to capture the spatial correlations of neighboring and relevant remote sensors. Another hybrid model, namely TYRE [31], uses GCN with gating and attention mechanisms as an intra-graph model and generates node embeddings at each time point to learn the temporal dependency.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Adaptive Graph Attention and Long Short-Term Memory-Based Networks for Traffic Prediction

Zhu,

Boada,

Boada

2024

Mathematics

View full text Add to dashboard Cite

While the increased availability of traffic data is allowing us to better understand urban mobility, research on data-driven and predictive modeling is also providing new methods for improving traffic management and reducing congestion. In this paper, we present a hybrid predictive modeling architecture, namely GAT-LSTM, by incorporating graph attention (GAT) and long short-term memory (LSTM) networks for handling traffic prediction tasks. In this architecture, GAT networks capture the spatial dependencies of the traffic network, LSTM networks capture the temporal correlations, and the Dayfeature component incorporates time and external information (such as day of the week, extreme weather conditions, holidays, etc.). A key attention block is designed to integrate GAT, LSTM, and the Dayfeature components as well as learn and assign weights to these different components within the architecture. This method of integration is proven effective at improving prediction accuracy, as shown by the experimental results obtained with the PeMS08 open dataset, and the proposed model demonstrates state-of-the-art performance in these experiments. Furthermore, the hybrid model demonstrates adaptability to dynamic traffic conditions, different prediction horizons, and various traffic networks.

show abstract