MGHCN: Multi-graph structures and hypergraph convolutional networks for traffic flow prediction
Accurate and timely traffic flow predictions are essential for effective traffic management and congestion reduction. However, most traditional prediction methods often fail to capture the complex dynamics and correlations within traffic flows due to insufficient processing of spatiotemporal data. S...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-01-01
|
| Series: | Alexandria Engineering Journal |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1110016824011773 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841524904310079488 |
|---|---|
| author | Xuanxuan Fan Kaiyuan Qi Dong Wu Haonan Xie Zhijian Qu Chongguang Ren |
| author_facet | Xuanxuan Fan Kaiyuan Qi Dong Wu Haonan Xie Zhijian Qu Chongguang Ren |
| author_sort | Xuanxuan Fan |
| collection | DOAJ |
| description | Accurate and timely traffic flow predictions are essential for effective traffic management and congestion reduction. However, most traditional prediction methods often fail to capture the complex dynamics and correlations within traffic flows due to insufficient processing of spatiotemporal data. Specifically, these methods struggle to integrate and analyze the multi-layered spatial and temporal interactions inherent in traffic data, leading to suboptimal prediction accuracy and robustness. To address this limitation, this paper presents a Multi-Graph Structures and Hypergraph Convolutional Network (MGHCN) that combines diverse graphs and hypergraphs. The MGHCN simplifies the predictive framework by integrating key components that improve its robustness and accuracy. One of the most critical components is the dual hypergraph structure, which captures edge correlations by converting traditional graph edges into hypergraph nodes. To better capture the spatiotemporal correlation of traffic data, a Graph Convolutional Network (GCN) is employed to analyze these hypergraphs in depth. Finally, a novel adjacency matrix and a dynamic graph module are used to accurately simulate interactions between spatiotemporal features, thereby enhancing the accuracy and robustness of predictions. Experimental validation on four distinct real-world traffic datasets shows that MGHCN outperforms existing state-of-the-art traffic prediction methods. |
| format | Article |
| id | doaj-art-8425f4dc87d8495cac9daf3f5e041bfa |
| institution | Kabale University |
| issn | 1110-0168 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Alexandria Engineering Journal |
| spelling | doaj-art-8425f4dc87d8495cac9daf3f5e041bfa2025-01-18T05:03:34ZengElsevierAlexandria Engineering Journal1110-01682025-01-01111221237MGHCN: Multi-graph structures and hypergraph convolutional networks for traffic flow predictionXuanxuan Fan0Kaiyuan Qi1Dong Wu2Haonan Xie3Zhijian Qu4Chongguang Ren5College of Computer Science and Technology, Shandong University of Technology, Zibo 255000, ChinaInspur (Jinan) Data Technology Co., Ltd, Jinan 250000, ChinaInspur (Jinan) Data Technology Co., Ltd, Jinan 250000, ChinaCollege of Computer Science and Technology, Shandong University of Technology, Zibo 255000, ChinaCollege of Computer Science and Technology, Shandong University of Technology, Zibo 255000, China; Corresponding author.College of Computer Science and Technology, Shandong University of Technology, Zibo 255000, China; Inspur (Jinan) Data Technology Co., Ltd, Jinan 250000, ChinaAccurate and timely traffic flow predictions are essential for effective traffic management and congestion reduction. However, most traditional prediction methods often fail to capture the complex dynamics and correlations within traffic flows due to insufficient processing of spatiotemporal data. Specifically, these methods struggle to integrate and analyze the multi-layered spatial and temporal interactions inherent in traffic data, leading to suboptimal prediction accuracy and robustness. To address this limitation, this paper presents a Multi-Graph Structures and Hypergraph Convolutional Network (MGHCN) that combines diverse graphs and hypergraphs. The MGHCN simplifies the predictive framework by integrating key components that improve its robustness and accuracy. One of the most critical components is the dual hypergraph structure, which captures edge correlations by converting traditional graph edges into hypergraph nodes. To better capture the spatiotemporal correlation of traffic data, a Graph Convolutional Network (GCN) is employed to analyze these hypergraphs in depth. Finally, a novel adjacency matrix and a dynamic graph module are used to accurately simulate interactions between spatiotemporal features, thereby enhancing the accuracy and robustness of predictions. Experimental validation on four distinct real-world traffic datasets shows that MGHCN outperforms existing state-of-the-art traffic prediction methods.http://www.sciencedirect.com/science/article/pii/S1110016824011773Traffic flow predictionGraph convolutional networkHypergraphMulti-graph structureSpatiotemporal analysis |
| spellingShingle | Xuanxuan Fan Kaiyuan Qi Dong Wu Haonan Xie Zhijian Qu Chongguang Ren MGHCN: Multi-graph structures and hypergraph convolutional networks for traffic flow prediction Alexandria Engineering Journal Traffic flow prediction Graph convolutional network Hypergraph Multi-graph structure Spatiotemporal analysis |
| title | MGHCN: Multi-graph structures and hypergraph convolutional networks for traffic flow prediction |
| title_full | MGHCN: Multi-graph structures and hypergraph convolutional networks for traffic flow prediction |
| title_fullStr | MGHCN: Multi-graph structures and hypergraph convolutional networks for traffic flow prediction |
| title_full_unstemmed | MGHCN: Multi-graph structures and hypergraph convolutional networks for traffic flow prediction |
| title_short | MGHCN: Multi-graph structures and hypergraph convolutional networks for traffic flow prediction |
| title_sort | mghcn multi graph structures and hypergraph convolutional networks for traffic flow prediction |
| topic | Traffic flow prediction Graph convolutional network Hypergraph Multi-graph structure Spatiotemporal analysis |
| url | http://www.sciencedirect.com/science/article/pii/S1110016824011773 |
| work_keys_str_mv | AT xuanxuanfan mghcnmultigraphstructuresandhypergraphconvolutionalnetworksfortrafficflowprediction AT kaiyuanqi mghcnmultigraphstructuresandhypergraphconvolutionalnetworksfortrafficflowprediction AT dongwu mghcnmultigraphstructuresandhypergraphconvolutionalnetworksfortrafficflowprediction AT haonanxie mghcnmultigraphstructuresandhypergraphconvolutionalnetworksfortrafficflowprediction AT zhijianqu mghcnmultigraphstructuresandhypergraphconvolutionalnetworksfortrafficflowprediction AT chongguangren mghcnmultigraphstructuresandhypergraphconvolutionalnetworksfortrafficflowprediction |