Attention Based Spatial-Temporal GCN with Kalman filter for Traffic Flow Prediction
Intelligent Transportation Systems (ITS) are becoming increasingly important as traditional traffic management systems struggle to handle the rapid growth of vehicles on the road. Accurate traffic prediction is a critical component of ITS, as it can help improve traffic management, avoid congest...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Universitas Indonesia
2023-10-01
|
| Series: | International Journal of Technology |
| Subjects: | |
| Online Access: | https://ijtech.eng.ui.ac.id/article/view/6646 |
| _version_ | 1797644353787658240 |
|---|---|
| author | Hatem Fahd Al-Selwi Azlan Abd. Aziz Fazly Bin Abas Aminuddin Kayani Noor Maizura Noor Siti Fatimah Abdul Razak |
| author_facet | Hatem Fahd Al-Selwi Azlan Abd. Aziz Fazly Bin Abas Aminuddin Kayani Noor Maizura Noor Siti Fatimah Abdul Razak |
| author_sort | Hatem Fahd Al-Selwi |
| collection | DOAJ |
| description | Intelligent
Transportation Systems (ITS) are becoming increasingly important as traditional
traffic management systems struggle to handle the rapid growth of vehicles on
the road. Accurate traffic prediction is a critical component of ITS, as it can
help improve traffic management, avoid congested roads, and allocate resources
more efficiently for connected vehicles. However, modeling traffic in a large
and interconnected road network is challenging because of its complex
spatio-temporal data. While classical statistics and machine learning methods
have been used for traffic prediction, they have limited ability to handle
complex traffic data, leading to unsatisfactory accuracy. In recent years, deep
learning methods, such as Recurrent Neural Networks (RNNs) and Convolutional
Neural Networks (CNNs), have shown superior capabilities for traffic
prediction. However, most CNN-based models are built for Euclidean
grid-structured data, while traffic road network data are irregular and better
formatted as graph-structured data. Graph Convolutional Neural Networks (GCNs)
have emerged to extend convolution operations to more general graph-structured
data. This paper reviews recent developments in traffic prediction using deep
learning, focusing on GCNs as a promising technique for handling irregular,
graph-structured traffic data. We also propose a novel GCN-based method that
leverages attention mechanisms to capture both local and long-range
dependencies in traffic data with Kalman Filter, and we demonstrate its
effectiveness through experiments on real-world datasets where the model
achieved around 5% higher accuracy compared to the original model. |
| first_indexed | 2024-03-11T14:30:23Z |
| format | Article |
| id | doaj.art-9728aaee01224dc0a766fe81c2658738 |
| institution | Directory Open Access Journal |
| issn | 2086-9614 2087-2100 |
| language | English |
| last_indexed | 2024-03-11T14:30:23Z |
| publishDate | 2023-10-01 |
| publisher | Universitas Indonesia |
| record_format | Article |
| series | International Journal of Technology |
| spelling | doaj.art-9728aaee01224dc0a766fe81c26587382023-10-31T10:11:52ZengUniversitas IndonesiaInternational Journal of Technology2086-96142087-21002023-10-011461299130810.14716/ijtech.v14i6.66466646Attention Based Spatial-Temporal GCN with Kalman filter for Traffic Flow PredictionHatem Fahd Al-Selwi0Azlan Abd. Aziz1Fazly Bin Abas2Aminuddin Kayani3Noor Maizura Noor4Siti Fatimah Abdul Razak5Faculty of Engineering and Technology, Multimedia University, Malacca 75450, MalaysiaFaculty of Engineering and Technology, Multimedia University, Malacca 75450, MalaysiaFaculty of Engineering and Technology, Multimedia University, Malacca 75450, MalaysiaSchool of Electrical and Computer Engineering, RMIT University, Melbourne, Victoria 3001, AustraliaDepartment of Computer Science, Faculty of Science and Technology, University Malaysia Terengganu Kuala Terengganu, 21030, Terengganu, Malaysia- Faculty of Information Science and Technology, Multimedia University, Melaka, MALAYSIAIntelligent Transportation Systems (ITS) are becoming increasingly important as traditional traffic management systems struggle to handle the rapid growth of vehicles on the road. Accurate traffic prediction is a critical component of ITS, as it can help improve traffic management, avoid congested roads, and allocate resources more efficiently for connected vehicles. However, modeling traffic in a large and interconnected road network is challenging because of its complex spatio-temporal data. While classical statistics and machine learning methods have been used for traffic prediction, they have limited ability to handle complex traffic data, leading to unsatisfactory accuracy. In recent years, deep learning methods, such as Recurrent Neural Networks (RNNs) and Convolutional Neural Networks (CNNs), have shown superior capabilities for traffic prediction. However, most CNN-based models are built for Euclidean grid-structured data, while traffic road network data are irregular and better formatted as graph-structured data. Graph Convolutional Neural Networks (GCNs) have emerged to extend convolution operations to more general graph-structured data. This paper reviews recent developments in traffic prediction using deep learning, focusing on GCNs as a promising technique for handling irregular, graph-structured traffic data. We also propose a novel GCN-based method that leverages attention mechanisms to capture both local and long-range dependencies in traffic data with Kalman Filter, and we demonstrate its effectiveness through experiments on real-world datasets where the model achieved around 5% higher accuracy compared to the original model.https://ijtech.eng.ui.ac.id/article/view/6646deep leaninggraphmachine learningtraffic prediction |
| spellingShingle | Hatem Fahd Al-Selwi Azlan Abd. Aziz Fazly Bin Abas Aminuddin Kayani Noor Maizura Noor Siti Fatimah Abdul Razak Attention Based Spatial-Temporal GCN with Kalman filter for Traffic Flow Prediction International Journal of Technology deep leaning graph machine learning traffic prediction |
| title | Attention Based Spatial-Temporal GCN with Kalman filter for Traffic Flow Prediction |
| title_full | Attention Based Spatial-Temporal GCN with Kalman filter for Traffic Flow Prediction |
| title_fullStr | Attention Based Spatial-Temporal GCN with Kalman filter for Traffic Flow Prediction |
| title_full_unstemmed | Attention Based Spatial-Temporal GCN with Kalman filter for Traffic Flow Prediction |
| title_short | Attention Based Spatial-Temporal GCN with Kalman filter for Traffic Flow Prediction |
| title_sort | attention based spatial temporal gcn with kalman filter for traffic flow prediction |
| topic | deep leaning graph machine learning traffic prediction |
| url | https://ijtech.eng.ui.ac.id/article/view/6646 |
| work_keys_str_mv | AT hatemfahdalselwi attentionbasedspatialtemporalgcnwithkalmanfilterfortrafficflowprediction AT azlanabdaziz attentionbasedspatialtemporalgcnwithkalmanfilterfortrafficflowprediction AT fazlybinabas attentionbasedspatialtemporalgcnwithkalmanfilterfortrafficflowprediction AT aminuddinkayani attentionbasedspatialtemporalgcnwithkalmanfilterfortrafficflowprediction AT noormaizuranoor attentionbasedspatialtemporalgcnwithkalmanfilterfortrafficflowprediction AT sitifatimahabdulrazak attentionbasedspatialtemporalgcnwithkalmanfilterfortrafficflowprediction |