DPSTCN: Dynamic Pattern-Aware Spatio-Temporal Convolutional Networks for Traffic Flow Forecasting
Accurate forecasting of multivariate traffic flow poses formidable challenges, primarily due to the ever-evolving spatio-temporal dynamics and intricate spatial heterogeneity, where the heterogeneity signifies that the correlations among locations are not just related to distance. However, few of th...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-12-01
|
| Series: | ISPRS International Journal of Geo-Information |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2220-9964/14/1/10 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832588380692545536 |
|---|---|
| author | Zeping Dou Danhuai Guo |
| author_facet | Zeping Dou Danhuai Guo |
| author_sort | Zeping Dou |
| collection | DOAJ |
| description | Accurate forecasting of multivariate traffic flow poses formidable challenges, primarily due to the ever-evolving spatio-temporal dynamics and intricate spatial heterogeneity, where the heterogeneity signifies that the correlations among locations are not just related to distance. However, few of the existing models are designed to fully and effectively integrate the above-mentioned features. To address these complexities head-on, this paper introduces a novel solution in the form of Dynamic Pattern-aware Spatio-Temporal Convolutional Networks (DPSTCN). Temporally, the model introduces a novel temporal module, containing a temporal convolutional network (TCN) enriched with an enhanced pattern-aware self-attention mechanism, adept at capturing temporal patterns, including local/global dependencies, dynamics, and periodicity. Spatially, the model constructs static and dynamic pattern-aware convolutions, leveraging geographical and area-functional information to effectively capture intricate spatial patterns, including dynamics and heterogeneity. Evaluations across four distinct traffic benchmark datasets consistently demonstrate the state-of-the-art capacity of our model compared to the existing eleven approaches, especially great improvements in RMSE (Root Mean Squared Error) value. |
| format | Article |
| id | doaj-art-e3fca0efe07242fda8476714d8986a17 |
| institution | Kabale University |
| issn | 2220-9964 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | ISPRS International Journal of Geo-Information |
| spelling | doaj-art-e3fca0efe07242fda8476714d8986a172025-01-24T13:34:58ZengMDPI AGISPRS International Journal of Geo-Information2220-99642024-12-011411010.3390/ijgi14010010DPSTCN: Dynamic Pattern-Aware Spatio-Temporal Convolutional Networks for Traffic Flow ForecastingZeping Dou0Danhuai Guo1College of Information Science and Technology, Beijing University of Chemical Technology, Beijing 100029, ChinaCollege of Information Science and Technology, Beijing University of Chemical Technology, Beijing 100029, ChinaAccurate forecasting of multivariate traffic flow poses formidable challenges, primarily due to the ever-evolving spatio-temporal dynamics and intricate spatial heterogeneity, where the heterogeneity signifies that the correlations among locations are not just related to distance. However, few of the existing models are designed to fully and effectively integrate the above-mentioned features. To address these complexities head-on, this paper introduces a novel solution in the form of Dynamic Pattern-aware Spatio-Temporal Convolutional Networks (DPSTCN). Temporally, the model introduces a novel temporal module, containing a temporal convolutional network (TCN) enriched with an enhanced pattern-aware self-attention mechanism, adept at capturing temporal patterns, including local/global dependencies, dynamics, and periodicity. Spatially, the model constructs static and dynamic pattern-aware convolutions, leveraging geographical and area-functional information to effectively capture intricate spatial patterns, including dynamics and heterogeneity. Evaluations across four distinct traffic benchmark datasets consistently demonstrate the state-of-the-art capacity of our model compared to the existing eleven approaches, especially great improvements in RMSE (Root Mean Squared Error) value.https://www.mdpi.com/2220-9964/14/1/10traffic forecastingintelligent transportation systemspatio-temporal data miningconvolutional networkattention mechanism |
| spellingShingle | Zeping Dou Danhuai Guo DPSTCN: Dynamic Pattern-Aware Spatio-Temporal Convolutional Networks for Traffic Flow Forecasting ISPRS International Journal of Geo-Information traffic forecasting intelligent transportation system spatio-temporal data mining convolutional network attention mechanism |
| title | DPSTCN: Dynamic Pattern-Aware Spatio-Temporal Convolutional Networks for Traffic Flow Forecasting |
| title_full | DPSTCN: Dynamic Pattern-Aware Spatio-Temporal Convolutional Networks for Traffic Flow Forecasting |
| title_fullStr | DPSTCN: Dynamic Pattern-Aware Spatio-Temporal Convolutional Networks for Traffic Flow Forecasting |
| title_full_unstemmed | DPSTCN: Dynamic Pattern-Aware Spatio-Temporal Convolutional Networks for Traffic Flow Forecasting |
| title_short | DPSTCN: Dynamic Pattern-Aware Spatio-Temporal Convolutional Networks for Traffic Flow Forecasting |
| title_sort | dpstcn dynamic pattern aware spatio temporal convolutional networks for traffic flow forecasting |
| topic | traffic forecasting intelligent transportation system spatio-temporal data mining convolutional network attention mechanism |
| url | https://www.mdpi.com/2220-9964/14/1/10 |
| work_keys_str_mv | AT zepingdou dpstcndynamicpatternawarespatiotemporalconvolutionalnetworksfortrafficflowforecasting AT danhuaiguo dpstcndynamicpatternawarespatiotemporalconvolutionalnetworksfortrafficflowforecasting |