A traffic prediction method for missing data scenarios: graph convolutional recurrent ordinary differential equation network
Abstract Traffic prediction plays an increasingly important role in intelligent transportation systems and smart cities. Both travelers and urban managers rely on accurate traffic information to make decisions about route selection and traffic management. Due to various factors, both human and natur...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Springer
2025-01-01
|
| Series: | Complex & Intelligent Systems |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/s40747-024-01768-7 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1823861519404761088 |
|---|---|
| author | Ming Jiang Zhiwei Liu Yan Xu |
| author_facet | Ming Jiang Zhiwei Liu Yan Xu |
| author_sort | Ming Jiang |
| collection | DOAJ |
| description | Abstract Traffic prediction plays an increasingly important role in intelligent transportation systems and smart cities. Both travelers and urban managers rely on accurate traffic information to make decisions about route selection and traffic management. Due to various factors, both human and natural, traffic data often contains missing values. Addressing the impact of missing data on traffic flow prediction has become a widely discussed topic in the academic community and holds significant practical importance. Existing spatiotemporal graph models typically rely on complete data, and the presence of missing values can significantly degrade prediction performance and disrupt the construction of dynamic graph structures. To address this challenge, this paper proposes a neural network architecture designed specifically for missing data scenarios—graph convolutional recurrent ordinary differential equation network (GCRNODE). GCRNODE combines recurrent networks based on ordinary differential equation (ODE) with spatiotemporal memory graph convolutional networks, enabling accurate traffic prediction and effective modeling of dynamic graph structures even in the presence of missing data. GCRNODE uses ODE to model the evolution of traffic flow and updates the hidden states of the ODE through observed data. Additionally, GCRNODE employs a data-independent spatiotemporal memory graph convolutional network to capture the dynamic spatial dependencies in missing data scenarios. The experimental results on three real-world traffic datasets demonstrate that GCRNODE outperforms baseline models in prediction performance under various missing data rates and scenarios. This indicates that the proposed method has stronger adaptability and robustness in handling missing data and modeling spatiotemporal dependencies. |
| format | Article |
| id | doaj-art-fce64cc60c2c4d8a9670e81bdd9238ee |
| institution | Kabale University |
| issn | 2199-4536 2198-6053 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Springer |
| record_format | Article |
| series | Complex & Intelligent Systems |
| spelling | doaj-art-fce64cc60c2c4d8a9670e81bdd9238ee2025-02-09T13:01:09ZengSpringerComplex & Intelligent Systems2199-45362198-60532025-01-0111212410.1007/s40747-024-01768-7A traffic prediction method for missing data scenarios: graph convolutional recurrent ordinary differential equation networkMing Jiang0Zhiwei Liu1Yan Xu2School of Internet Economics and Business, Fujian University of TechnologySchool of Transportation, Fujian University of TechnologySchool of Transportation, Fujian University of TechnologyAbstract Traffic prediction plays an increasingly important role in intelligent transportation systems and smart cities. Both travelers and urban managers rely on accurate traffic information to make decisions about route selection and traffic management. Due to various factors, both human and natural, traffic data often contains missing values. Addressing the impact of missing data on traffic flow prediction has become a widely discussed topic in the academic community and holds significant practical importance. Existing spatiotemporal graph models typically rely on complete data, and the presence of missing values can significantly degrade prediction performance and disrupt the construction of dynamic graph structures. To address this challenge, this paper proposes a neural network architecture designed specifically for missing data scenarios—graph convolutional recurrent ordinary differential equation network (GCRNODE). GCRNODE combines recurrent networks based on ordinary differential equation (ODE) with spatiotemporal memory graph convolutional networks, enabling accurate traffic prediction and effective modeling of dynamic graph structures even in the presence of missing data. GCRNODE uses ODE to model the evolution of traffic flow and updates the hidden states of the ODE through observed data. Additionally, GCRNODE employs a data-independent spatiotemporal memory graph convolutional network to capture the dynamic spatial dependencies in missing data scenarios. The experimental results on three real-world traffic datasets demonstrate that GCRNODE outperforms baseline models in prediction performance under various missing data rates and scenarios. This indicates that the proposed method has stronger adaptability and robustness in handling missing data and modeling spatiotemporal dependencies.https://doi.org/10.1007/s40747-024-01768-7Intelligent transportationTraffic predictionMissing dataGraph convolutional neural networkNeural ordinary differential equation |
| spellingShingle | Ming Jiang Zhiwei Liu Yan Xu A traffic prediction method for missing data scenarios: graph convolutional recurrent ordinary differential equation network Complex & Intelligent Systems Intelligent transportation Traffic prediction Missing data Graph convolutional neural network Neural ordinary differential equation |
| title | A traffic prediction method for missing data scenarios: graph convolutional recurrent ordinary differential equation network |
| title_full | A traffic prediction method for missing data scenarios: graph convolutional recurrent ordinary differential equation network |
| title_fullStr | A traffic prediction method for missing data scenarios: graph convolutional recurrent ordinary differential equation network |
| title_full_unstemmed | A traffic prediction method for missing data scenarios: graph convolutional recurrent ordinary differential equation network |
| title_short | A traffic prediction method for missing data scenarios: graph convolutional recurrent ordinary differential equation network |
| title_sort | traffic prediction method for missing data scenarios graph convolutional recurrent ordinary differential equation network |
| topic | Intelligent transportation Traffic prediction Missing data Graph convolutional neural network Neural ordinary differential equation |
| url | https://doi.org/10.1007/s40747-024-01768-7 |
| work_keys_str_mv | AT mingjiang atrafficpredictionmethodformissingdatascenariosgraphconvolutionalrecurrentordinarydifferentialequationnetwork AT zhiweiliu atrafficpredictionmethodformissingdatascenariosgraphconvolutionalrecurrentordinarydifferentialequationnetwork AT yanxu atrafficpredictionmethodformissingdatascenariosgraphconvolutionalrecurrentordinarydifferentialequationnetwork AT mingjiang trafficpredictionmethodformissingdatascenariosgraphconvolutionalrecurrentordinarydifferentialequationnetwork AT zhiweiliu trafficpredictionmethodformissingdatascenariosgraphconvolutionalrecurrentordinarydifferentialequationnetwork AT yanxu trafficpredictionmethodformissingdatascenariosgraphconvolutionalrecurrentordinarydifferentialequationnetwork |