An effective dynamic spatiotemporal framework with external features information for traffic prediction

Wang, Jichen; Zhu, Weiguo; Sun, Yeneng; Tian, Chunzi

doi:10.1007/s10489-020-02043-1

Cited by 26 publications

(5 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The system relies on a mapping process where the raw spatiotemporal trajectories of the users are translated into grid-based tessellations of the urban setting. Wang et al [30] extended this bidirectional LSTM model with an attention mechanism to forecast urban traffic conditions in peak hours using external factors such as weather conditions. The authors tested their proposal using taxi and bike datasets from New York.…”

Section: Related Workmentioning

confidence: 99%

“…The mechanism was tested using different location data at different granularities in three cities, New York, Beijing, and Shanghai. Wang et al [30] proposed the combination of a GRU with a CNN along with an attention mechanism to predict the traffic speed on certain roads of Paris. This proposal can detect traffic trends depending on the day of the week and without using past events.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Nation-wide human mobility prediction based on graph neural networks

Terroso-Sáenz

Muñoz

2021

Appl Intell

View full text Add to dashboard Cite

Nowadays, the anticipation of human mobility flow has important applications in many domains ranging from urban planning to epidemiology. Because of the high predictability of human movements, numerous successful solutions to perform such forecasting have been proposed. However, most focus on predicting human displacements on an intra-urban spatial scale. This study proposes a predictor for nation-wide mobility that allows anticipating inter-urban displacements at larger spatial granularity. For this goal, a Graph Neural Network (GNN) was used to consider the latent relationships among large geographical regions. The solution has been evaluated with an open dataset including trips throughout the country of Spain and the current weather conditions. The results indicate a high accuracy in predicting the number of trips for multiple time horizons, and more important, they show that our proposal only needs a single model for processing all the mobility areas in the dataset, whereas other techniques require a different model for each area under study.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Nation-wide human mobility prediction based on graph neural networks

Terroso-Sáenz

Muñoz

2021

Appl Intell

View full text Add to dashboard Cite

show abstract

“…Ref. [43] proposed a model with bidirectional long short-term memory (LSTM) and a complex attention mechanism to predict the urban traffic volume, combined with weather conditions and event information as external features to further improve the prediction precision. A traffic graph convolution operator was proposed in [44] in order to extract the local features and combine the physical features of the road network.…”

Section: Traffic Speed Predictionmentioning

confidence: 99%

The Effect of Crowdsourced Police Enforcement Data on Traffic Speed: A Case Study of The Netherlands

Liu,

Feng

2023

Applied Sciences

View full text Add to dashboard Cite

The proliferation of smartphones and internet connectivity has provided the opportunity to use crowdsourced data in traffic management. Nowadays, many people use navigation apps such as Google Maps, Waze, and Flitsmeister to obtain real-time travel information and provide feedback on road conditions, such as reporting police speed checks. As an accurate traffic speed prediction is of great significance for road users and traffic managers, different models have been proposed and widely used to predict traffic speed considering the spatio-temporal dependence of traffic data and external factors such as the weather, accidents and points of interest. This study investigates the impact of crowdsourced data about police enforcement from navigation apps on traffic speed. In addition, we examine whether the police enforcement report affects the accuracy of the deep learning prediction model. The authors extract crowdsourced police enforcement information from navigation apps, collect the corresponding historical traffic speed data, and predict traffic speed in several corridors in The Netherlands using a GCN-GRU traffic speed prediction model. The results show that the crowdsourced data for police enforcement cause the average vehicle speed to drop between 1 [km/h] and 3 [km/h] when passing the road segments marked with police activity. Moreover, the prediction performance of the GCN-GRU model during the periods without police enforcement is better than the periods with reported police activity, showing that police speed check reports can decrease the accuracy of speed prediction models.

show abstract

“…It can be noticed that there is more focus in these studies on exploring the spatial and temporal traffic features when predicting traffic conditions using the BiLSTM model [33,34,35,36,37]. However, few studies have explored the feasibility of this type of model to be validated or transferred (without retraining) to an independent dataset from a different freeway [38] or in the case of this paper, validate the model against future traffic scenarios where the demand is expected to increased to up to 80% in the future.…”

Section: Literature Reviewmentioning

confidence: 99%

Development and Evaluation of Bidirectional LSTM Freeway Traffic Forecasting Models Using Simulation Data

Abduljabbar

Dia

Tsai

2021

Preprint

View full text Add to dashboard Cite

Long short-term memory (LSTM) models provide high predictive performance through their ability to recognize longer sequences of time series data. More recently, bidirectional deep learning models (BiLSTM) have extended the LSTM capabilities by training the input data twice in forward and backward directions. In this paper, BiLSTM short term traffic forecasting models have been developed and evaluated using data from a calibrated micro-simulation model for a congested freeway in Melbourne, Australia. The simulation model was extensively calibrated and validated to a high degree of accuracy using field data collected from 55 detectors on the freeway. The base year simulation model was then used to generate loop detector data including speed, flow and occupancy which were used to develop and compare a number of LSTM models for short-term traffic prediction up to 60 minutes into the future. The modelling results showed that BiLSTM outperformed other predictive models for multiple prediction horizons for base year conditions. The simulation model was then adapted for future year scenarios where the traffic demand was increased by 25-100 percent to reflect potential future increases in traffic demands. The results showed superior performance of BiLSTM for multiple prediction horizons for all traffic variables.

show abstract

An effective dynamic spatiotemporal framework with external features information for traffic prediction

Cited by 26 publications

References 25 publications

Nation-wide human mobility prediction based on graph neural networks

Nation-wide human mobility prediction based on graph neural networks

The Effect of Crowdsourced Police Enforcement Data on Traffic Speed: A Case Study of The Netherlands

Development and Evaluation of Bidirectional LSTM Freeway Traffic Forecasting Models Using Simulation Data

Contact Info

Product

Resources

About