Spatial temporal incidence dynamic graph neural networks for traffic flow forecasting

Peng, Hao; Wang, Hongfei; Du, Bowen; Bhuiyan, Zakirul Alam; Ma, Hongyuan; Liu, Jianwei; Wang, Lihong; Yang, Zeyu; Du, Linfeng; Wang, Senzhang; Yu, Philip S.

doi:10.1016/j.ins.2020.01.043

Cited by 261 publications

(81 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Traffic flow is the most fundamental criterion of understanding road capacity and traffic congestion, and it is divided into long-term and short-term predictions [42]. It is an essential measurement for travel navigation decisions [40], transportation management [41], smart city planning [42], and others. Most of the research conducted in this specific area aims to propose a better traffic handling mechanism by making full use of the various source of data, such as GPS, the incoming flow of vehicle, the outgoing flow of vehicle, even meteorological data, including weather, temperature, and wind speed [40], [41].…”

Section: A Tfa Attributesmentioning

confidence: 99%

“…Peng et al [41] propose a convolutional NN (CNN) with a deep learning model to generate traffic flow forecasting by using historical data of subway, taxi, and bus in Beijing. Evaluation and comparison among different methods that used the same dataset (TaxiBJ) are conducted.…”

Section: ) Neural Networkmentioning

confidence: 99%

“…Few TFA related studies discuss AI contribution to DF framework when integrating with huge data from various data sources [9], [62], [63], [100], [88]. AI-related techniques have improved few areas of DF implementation in TFA related study, such as traffic flow prediction [41], [88], [106], [125], pedestrian detection module [126], traffic data classification [127], traffic speed prediction [38] and missing value estimation. Figure 4 DL methods are known for the ability to manage the fusion process of heterogeneous data from multiple sources and robust enough to handle substantial unsupervised data from those multiple heterogeneous sources [69].…”

Section: ) Hybrid Algorithmmentioning

confidence: 99%

“…It is used in traffic state estimation [12], [86] and vehicle turning ratio estimation [95]. NN is best served as a feature level fusion technique to produce flow prediction [40], [41], and filtering pedestrian detection [99]. DS works best as a classifier at decision level fusion to detect lane change activity [45], detect misbehaved clusters of vehicles [102], and traffic state estimation [70].…”

Section: A Analysismentioning

confidence: 99%

See 3 more Smart Citations

Review of Data Fusion Methods for Real-Time and Multi-Sensor Traffic Flow Analysis

et al. 2021

View full text Add to dashboard Cite

Recently, development in intelligent transportation systems (ITS) requires the input of various kinds of data in real-time and from multiple sources, which imposes additional research and application challenges. Ongoing studies on Data Fusion (DF) have produced significant improvement in ITS and manifested an enormous impact on its growth. This paper reviews the implementation of DF methods in ITS to facilitate traffic flow analysis (TFA) and solutions that entail the prediction of various traffic variables such as driving behavior, travel time, speed, density, incident, and traffic flow. It attempts to identify and discuss real-time and multi-sensor data sources that are used for various traffic domains, including road/highway management, traffic states estimation, and traffic controller optimization. Moreover, it attempts to associate abstractions of data level fusion, feature level fusion, and decision level fusion on DF methods to better understand the role of DF in TFA and ITS. Consequently, the main objective of this paper is to review DF methods used for real-time and multi-sensor (heterogeneous) TFA studies. The review outcomes are (i) a guideline of constructing DF methods which involve preprocessing, filtering, decision, and evaluation as core steps, (ii) a description of the recent DF algorithms or methods that adopt real-time and multi-sensor sources data and the impact of these data sources on the improvement of TFA, (iii) an examination of the testing and evaluation methodologies and the popular datasets and (iv) an identification of several research gaps, some current challenges, and new research trends.INDEX TERMS Intelligent transportation systems, traffic flow analysis, data fusion; real-time processing, multi-sensor, heterogeneous data, machine learning.

show abstract

Section: A Tfa Attributesmentioning

confidence: 99%

Section: ) Neural Networkmentioning

confidence: 99%

Section: ) Hybrid Algorithmmentioning

confidence: 99%

Section: A Analysismentioning

confidence: 99%

See 2 more Smart Citations

Review of Data Fusion Methods for Real-Time and Multi-Sensor Traffic Flow Analysis

et al. 2021

View full text Add to dashboard Cite

show abstract

“…In contrast, there is limited research on graph-based documentation representation [11]. Compared to bag of words/phrases representation, graph-based document modeling can preserve local sequential, non-consecutive, and long-distance semantics, and consequently provides improved classification accuracy [12], [13].…”

Section: Introductionmentioning

confidence: 99%

Hierarchical Graph Transformer-Based Deep Learning Model for Large-Scale Multi-Label Text Classification

Gong

Teng

et al. 2020

IEEE Access

Self Cite

View full text Add to dashboard Cite

Traditional methods of multi-label text classification, particularly deep learning, have achieved remarkable results. However, most of these methods use word2vec technology to represent sequential text information, while ignoring the logic and internal hierarchy of the text itself. Although these approaches can learn the hypothetical hierarchy and logic of the text, it is unexplained. In addition, the traditional approach treats labels as independent individuals and ignores the relationships between them, which not only does not reflect reality but also causes significant loss of semantic information. In this paper, we propose a novel Hierarchical Graph Transformer based deep learning model for large-scale multi-label text classification. We first model the text into a graph structure that can embody the different semantics of the text and the connections between them. We then use a multi-layer transformer structure with a multi-head attention mechanism at the word, sentence, and graph levels to fully capture the features of the text and observe the importance of the separate parts. Finally, we use the hierarchical relationship of the labels to generate the representation of the labels, and design a weighted loss function based on the semantic distances of the labels. Extensive experiments conducted on three benchmark datasets demonstrated that the proposed model can realistically capture the hierarchy and logic of text and improve performance compared with the state-of-the-art methods. INDEX TERMS Multi-label text classification, graph modeling, graph transformer, deep learning.

show abstract

Modeling Urban Traffic by Means of Traffic Density Data: Istanbul Case

Yaman

Başeğmez

Sezer

2020

Advances in Intelligent Systems and Computing

View full text Add to dashboard Cite

Spatial temporal incidence dynamic graph neural networks for traffic flow forecasting

Cited by 261 publications

References 20 publications

Review of Data Fusion Methods for Real-Time and Multi-Sensor Traffic Flow Analysis

Review of Data Fusion Methods for Real-Time and Multi-Sensor Traffic Flow Analysis

Hierarchical Graph Transformer-Based Deep Learning Model for Large-Scale Multi-Label Text Classification

Modeling Urban Traffic by Means of Traffic Density Data: Istanbul Case

Contact Info

Product

Resources

About