Managing Spatial Graph Dependencies in Large Volumes of Traffic Data for Travel-Time Prediction

Salamanis, Athanasios; Kehagias, Dionysios; Filelis‐Papadopoulos, Christos K.; Tzovaras, Dimitrios; Gravvanis, George A.

doi:10.1109/tits.2015.2488593

Cited by 77 publications

(29 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The classical specification of VAR, the most popular multivariate time series model, is rarely applied to large spatial segments, due to its enormous number of parameters and the potential problem of overfitting. Thus, several sparse specifications of VAR were suggested based on road connectivity [33], cross-correlation [34], adaptive LASSO [35], among others. Drawing a parallel with video processing methodologies, the VAR model in spatial settings represents a linear pixel-specific data-driven spatial kernel, while the sparse VAR model specifications are based on a graph of spatiotemporal dependencies.…”

Section: Spatiotemporal Urban Traffic Forecasting Methodologymentioning

confidence: 99%

Transfer Learning: Video Prediction and Spatiotemporal Urban Traffic Forecasting

Pavlyuk

2020

Algorithms

View full text Add to dashboard Cite

Transfer learning is a modern concept that focuses on the application of ideas, models, and algorithms, developed in one applied area, for solving a similar problem in another area. In this paper, we identify links between methodologies in two fields: video prediction and spatiotemporal traffic forecasting. The similarities of the video stream and citywide traffic data structures are discovered and analogues between historical development and modern states of the methodologies are presented and discussed. The idea of transferring video prediction models to the urban traffic forecasting domain is validated using a large real-world traffic data set. The list of transferred techniques includes spatial filtering by predefined kernels in combination with time series models and spectral graph convolutional artificial neural networks. The obtained models’ forecasting performance is compared to the baseline traffic forecasting models: non-spatial time series models and spatially regularized vector autoregression models. We conclude that the application of video prediction models and algorithms for urban traffic forecasting is effective both in terms of observed forecasting accuracy and development, and training efforts. Finally, we discuss problems and obstacles of transferring methodologies and present potential directions for further research.

show abstract

Section: Spatiotemporal Urban Traffic Forecasting Methodologymentioning

confidence: 99%

Transfer Learning: Video Prediction and Spatiotemporal Urban Traffic Forecasting

Pavlyuk

2020

Algorithms

View full text Add to dashboard Cite

show abstract

“…Such composed inputs do not represent the STN structure, but do include its most important aspects. PCA-EVD was used in nine studies as a preliminary step for different forecasting models: neural networks [69,70], support vector regression [71][72][73][74][75], Bayesian networks [76], and random forests [77]. PCA-EVD has also been used as a method for tensor decomposition [78].…”

Section: Class 5: Dimension Reduction Methodsmentioning

confidence: 99%

Feature selection and extraction in spatiotemporal traffic forecasting: a systematic literature review

Pavlyuk

2019

Eur. Transp. Res. Rev.

View full text Add to dashboard Cite

A spatiotemporal approach that simultaneously utilises both spatial and temporal relationships is gaining scientific interest in the field of traffic flow forecasting. Accurate identification of the spatiotemporal structure (dependencies amongst traffic flows in space and time) plays a critical role in modern traffic forecasting methodologies, and recent developments of data-driven feature selection and extraction methods allow the identification of complex relationships. This paper systematically reviews studies that apply feature selection and extraction methods for spatiotemporal traffic forecasting. The reviewed bibliographic database includes 211 publications and covers the period from early 1984 to March 2018. A synthesis of bibliographic sources clarifies the advantages and disadvantages of different feature selection and extraction methods for learning the spatiotemporal structure and discovers trends in their applications. We conclude that there is a clear need for development of comprehensive guidelines for selecting appropriate spatiotemporal feature selection and extraction methods for urban traffic forecasting.

show abstract

“…In this work, a large amount of historical data was processed (six million trips from 20 thousand taxis and 400 thousand POI records in Beijing) with the risk of becoming irrelevant due to fast changes of the traffic context in time. The authors of work in [ 43 ] present a STARIMA based approach that efficiently predicts travel time using large volumes of traffic data information in Berlin and Thessaloniki. A communications-oriented perspective on traffic management systems for smart cities is discussed in [ 44 ] with main focus on short-term traffic forecasting.…”

Section: Related Workmentioning

confidence: 99%

Scalable Data Model for Traffic Congestion Avoidance in a Vehicle to Cloud Infrastructure

Stan

Suciu

Potolea

2021

Sensors

View full text Add to dashboard Cite

Traffic congestion experience in urban areas has negative impact on our daily lives by consuming our time and resources. Intelligent Transportation Systems can provide the necessary infrastructure to mitigate such challenges. In this paper, we propose a novel and scalable solution to model, store and control traffic data based on range query data structures (K-ary Interval Tree and K-ary Entry Point Tree) which allows data representation and handling in a way that better predicts and avoids traffic congestion in urban areas. Our experiments, validation scenarios, performance measurements and solution assessment were done on Brooklyn, New York traffic congestion simulation scenario and shown the validity, reliability, performance and scalability of the proposed solution in terms of time spent in traffic, run-time and memory usage. The experiments on the proposed data structures simulated up to 10,000 vehicles having microseconds time to access traffic information and below 1.5 s for congestion free route generation in complex scenarios. To the best of our knowledge, this is the first scalable approach that can be used to predict urban traffic and avoid congestion through range query data structure traffic modelling.

show abstract

Managing Spatial Graph Dependencies in Large Volumes of Traffic Data for Travel-Time Prediction

Cited by 77 publications

References 21 publications

Transfer Learning: Video Prediction and Spatiotemporal Urban Traffic Forecasting

Transfer Learning: Video Prediction and Spatiotemporal Urban Traffic Forecasting

Feature selection and extraction in spatiotemporal traffic forecasting: a systematic literature review

Scalable Data Model for Traffic Congestion Avoidance in a Vehicle to Cloud Infrastructure

Contact Info

Product

Resources

About