Dynamic Correlation Adjacency Matrix Based Graph Neural Network for Traffic Flow Prediction
Article
Article Title | Dynamic Correlation Adjacency Matrix Based Graph Neural Network for Traffic Flow Prediction |
---|---|
ERA Journal ID | 34304 |
Article Category | Article |
Authors | Gu, Junhua, Jia, Zhihao, Cai, Taotao, Song, Xiangyu and Mahmood, Adnan |
Journal Title | Sensors |
Journal Citation | 23 (6) |
Article Number | 2897 |
Number of Pages | 17 |
Year | Mar 2023 |
Publisher | MDPI AG |
Place of Publication | Switzerland |
ISSN | 1424-8220 |
1424-8239 | |
Digital Object Identifier (DOI) | https://doi.org/10.3390/s23062897 |
Web Address (URL) | https://www.mdpi.com/1424-8220/23/6/2897 |
Abstract | Modeling complex spatial and temporal dependencies in multivariate time series data is crucial for traffic forecasting. Graph convolutional networks have proved to be effective in predicting multivariate time series. Although a predefined graph structure can help the model converge to good results quickly, it also limits the further improvement of the model due to its stationary state. In addition, current methods may not converge on some datasets due to the graph structure of these datasets being difficult to learn. Motivated by this, we propose a novel model named Dynamic Correlation Graph Convolutional Network (DCGCN) in this paper. The model can construct adjacency matrices from input data using a correlation coefficient; thus, dynamic correlation graph convolution is used for capturing spatial dependencies. Meanwhile, gated temporal convolution is used for modeling temporal dependencies. Finally, we performed extensive experiments to evaluate the performance of our proposed method against ten existing well-recognized baseline methods using two original and four public datasets. |
Keywords | graph neural networks; dynamic adjacency matrix; multivariate time series; traffic prediction |
Contains Sensitive Content | Does not contain sensitive content |
ANZSRC Field of Research 2020 | 4605. Data management and data science |
Byline Affiliations | Hebei University of Technology, China |
School of Mathematics, Physics and Computing | |
Swinburne University of Technology | |
Macquarie University |
https://research.usq.edu.au/item/yzz2v/dynamic-correlation-adjacency-matrix-based-graph-neural-network-for-traffic-flow-prediction
Download files
35
total views19
total downloads1
views this month0
downloads this month