T-Rank: Graph Data Analytics for Urban Traffic Modeling

Evaluating specific points in urban passages and identifying the factors causing traffic jams are of particular importance. In this paper, we seek for a mathematical model of urban traffic that identifies the factors affecting traffic and evaluating their impact. Such models are capable of predictin...

Full description

Saved in:

Bibliographic Details
Published in:	2021 11th International Conference on Computer Engineering and Knowledge (ICCKE) pp. 387 - 392
Main Authors:	Safarpour, Alireza, Gholampour, Iman, Fard, Amirhossain Aghazadeh, Karbasi, Seyed Mohammad
Format:	Conference Proceeding
Language:	English
Published:	IEEE 28-10-2021
Subjects:	Analytical models Cameras Data analysis Intelligent Transport Systems Knowledge engineering Massive Data Analytics Mathematical analysis Prediction algorithms Predictive models Spark and Hadoop urban Traffic
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	Evaluating specific points in urban passages and identifying the factors causing traffic jams are of particular importance. In this paper, we seek for a mathematical model of urban traffic that identifies the factors affecting traffic and evaluating their impact. Such models are capable of predicting the impact of the latent factors on the urban traffic. Our model uses graph evaluation algorithms, similar to ones used in web-page ranking. The model validation is done based on the data gathered by ANPR 1 cameras that detect the plate numbers of the vehicles in the traffic zones. The data analytics in Apache Spark environment shows that our model archives an accuracy of more than 80% in daily traffic density prediction. The model also predicts the ratio of the vehicles that park in the monitored zone with respect to the ones that pass through the zone accurately.
ISSN:	2643-279X
DOI:	10.1109/ICCKE54056.2021.9721454