Hybrid deep learning models for traffic prediction in large-scale road networks

Ge, Zheng; Chai, Wei Koong; Duanmu, Jing‐Lin; Katos, Vasilios

doi:10.1016/j.inffus.2022.11.019

Cited by 43 publications

(5 citation statements)

References 99 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Meanwhile, recent advancements in artificial intelligence have given rise to new machine learning (ML) and deep learning (DL) models developed for traffic prediction. We refer readers to Zheng et al [42] and the references therein for an overview of the evolution of the research along this direction from simple ML approaches (e.g., K-nearest neighbour (KNN) algorithm [43], Artificial Neural Network (ANN) [44], [45] to DL models (e.g., Recurrent Neural Networks (RNN) and its variants (Long-Short Term Memory (LSTM) and Gated Recurrent Unit (GRU) [46], Convolutional Neural Network (CNN) [47], and Graph Convolutional Neural Network (GCN) [48]. Here, rather than understanding the congestion spreading, the focus is on predicting the state of the road traffic in the next prediction horizon (e.g., could be set to 5-min or 10-min in the future) based on historical data.…”

Section: Related Workmentioning

confidence: 99%

Modeling Traffic Congestion Spreading Using a Topology-Based SIR Epidemic Model

Kozhabek,

Chai,

Zheng

2024

IEEE Access

Self Cite

View full text Add to dashboard Cite

The continuous urbanisation and increase in vehicle ownership have increasingly exacerbated traffic congestion problems. In this paper, we advocate the use of epidemic theory to model the spreading of traffic congestion in urban cities. Specifically, we use the Susceptible-Infected-Recovered (SIR) model but propose to explicitly consider the road network structure in the model to understand the contagion process of road congestion. This departs from the classical SIR model where homogeneous mixing based on the law of mass action is assumed. For this purpose, we adopt the N-intertwined modeling framework for the SIR model based on continuous-time Markov chain analysis. In our evaluation, we used two real-world traffic datasets collected in California and Los Angeles. We compare our results against both classical and average-degreebased SIR models. Our results show better agreement between the model and actual congestion conditions and shed light on how congestion propagates across a road network. We see the potential application of insights gained from this work on the development of traffic congestion mitigation strategies.INDEX TERMS Congestion spread modeling, epidemics, SIR model, topology,traffic congestion, urban road networks.

show abstract

Section: Related Workmentioning

confidence: 99%

Modeling Traffic Congestion Spreading Using a Topology-Based SIR Epidemic Model

Kozhabek,

Chai,

Zheng

2024

IEEE Access

Self Cite

View full text Add to dashboard Cite

show abstract

“…Zheng et al developed a hybrid deep learning model using CNN and LSTM modules to extract temporal and spatial properties of traffic flow data [15]. An attentional mechanism was developed by assigning different weights to recipe flow data at different times.…”

Section: Essien Et Al Propose a Deep Learning-based Trafficmentioning

confidence: 99%

Kentsel Trafik Tahminine Yönelik Derin Öğrenme Tabanlı Verimli Bir Hibrit Model

Utku

2023

Bilişim Teknolojileri Dergisi

View full text Add to dashboard Cite

The traffic density problem has become one of the most important problems of urban life. The time and fuel spent due to traffic density is a significant loss for vehicle users and countries. Applications developed to reduce the time spent in traffic cannot make successful predictions about long-term traffic density. Traffic data obtained from cameras, sensors and mobile devices highlights the use of artificial intelligence technologies in order to solve the traffic management problem. In this study, a hybrid prediction model has been proposed by using CNN and RNN models for traffic density prediction. The proposed hybrid model has been tested using LR, RF, SVM, MLP, CNN, RNN and LSTM and Istanbul's traffic data for 2020. Experimental results showed that the proposed hybrid model has more successful results than the compared models. The proposed model has 0.929 R2 in the prediction of the number of vehicles passing through the junction, and 0.934 R2 in the prediction of the average speed of the vehicles passing through the junction.

show abstract

“…Long Short Term Memory (LSTM) [23] networks, for example, exhibited significant prowess in understanding long-term patterns [24]. Conversely, Temporal Convolutional Networks (TCN)s [25], have been heralded for their efficiency in capturing short-term temporal relations, especially in Spatio-Temporal series where spatial interactions are pivotal.…”

Section: Background and Data Exploration A Ai For Air Qualitymentioning

confidence: 99%

Data Driven Forecasting Models for Urban Air Pollution: MoreAir Case Study

Berkani,

Gryech,

Ghogho

et al. 2023

IEEE Access

View full text Add to dashboard Cite

Artificial Intelligence has the potential to contribute to sustainable cities, life on land, and climate action. Specifically, data-driven AI models can analyze large, interconnected databases to develop joint environmental actions. Air quality plays a pivotal role in both climate action and the development of sustainable cities, but developing countries face challenges due to insufficient monitoring stations and limited access to air quality data sets. This study builds upon the MoreAir project, which established a low-cost air pollution monitoring system and provided the first air quality data set from Morocco. We first exploit and delve into the details of the obtained dataset. Subsequently, we conduct a multi-level comparison of data-driven forecasting models, specifically focusing on short-term forecasting of Particulate Matter concentrations. Four forecasting frameworks are explored, using different combinations of exogenous data and spatio-temporal information. Our findings highlight that Machine Learning models, particularly LightGBM and CatBoost, outperform other models. Overall, our study demonstrates that the inclusion of the spatial dimension along with the diverse exogenous features enhances the models' predictive performance, and provides valuable insights.

show abstract

Hybrid deep learning models for traffic prediction in large-scale road networks

Cited by 43 publications

References 99 publications

Modeling Traffic Congestion Spreading Using a Topology-Based SIR Epidemic Model

Modeling Traffic Congestion Spreading Using a Topology-Based SIR Epidemic Model

Kentsel Trafik Tahminine Yönelik Derin Öğrenme Tabanlı Verimli Bir Hibrit Model

Data Driven Forecasting Models for Urban Air Pollution: MoreAir Case Study

Contact Info

Product

Resources

About