Embedding Traffic Network Characteristics Using Tensor for Improved Traffic Prediction

被引:25
作者
Bhanu, Manish [1 ]
Mendes-Moreira, Joao [2 ,3 ]
Chandra, Joydeep [1 ]
机构
[1] Indian Inst Technol, Dept Comp Sci & Engn, Patna 801106, Bihar, India
[2] Univ Porto, Fac Engn, Dept Informat Engn, P-4099002 Porto, Portugal
[3] INESC TEC, LIAAD, P-4200465 Porto, Portugal
关键词
Tensile stress; Urban areas; Matrix decomposition; Forecasting; Public transportation; Predictive models; Traffic prediction; tensor decomposition; network characteristics; reciprocity; DIMENSIONALITY REDUCTION; DECOMPOSITIONS; MOBILITY; MODEL;
D O I
10.1109/TITS.2020.2984175
中图分类号
TU [建筑科学];
学科分类号
0813 ;
摘要
Techniques for using multi-way traffic patterns for traffic prediction is gaining importance. One of the possible techniques for representing the multi-way traffic patterns is tensors. Tensor decomposition is used to generate low-rank approximations of the original tensor that is subsequently used for traffic volume prediction. However, the existing tensor-based approaches do not consider certain important mutual relationships among the locations like temporal traffic reciprocity that can improve the prediction accuracy. In this paper, we introduce TeDCaN, a "Tensor Decomposition method with Characteristic Network" constraints that generate low rank approximations of the original tensor considering the traffic reciprocity at different pair of locations. Investigations using large traffic datasets from 2 different cities reveal that the prediction accuracy of TeDCaN considerably outperforms several state-of-art baselines for cases when complete traffic data is available as well as situations when a certain fraction of the data is missing - a likely scenario in many real datasets. We discover that TeDCaN achieves around 20% reduction in the RMSE scores as compared to the baselines. TeDCaN is applicable in many operations on such a big traffic network where the existing models would either be inapplicable or hard to perform. As one of the major yields, TeDCaN generates a "reduced dimensional network embedding" that captures the similarity of the nodes considering the traffic volume as well as the reciprocity of traffic between the nodes.
引用
收藏
页码:3359 / 3371
页数:13
相关论文
共 41 条
[1]  
[Anonymous], 2017, ABS170904875 CORR
[2]  
[Anonymous], 2011, P AS PAC POW EN ENG
[3]  
Azzouni A., 2017, ARXIV170505690
[4]   Travel Time Forecasting and Dynamic Origin-Destination Estimation for Freeways Based on Bluetooth Traffic Monitoring [J].
Barcelo, Jaume ;
Montero, Lidin ;
Marques, Laura ;
Carmona, Carlos .
TRANSPORTATION RESEARCH RECORD, 2010, (2175) :19-27
[5]  
Battaglino C., 2017, ARXIV170106600
[6]   Laplacian eigenmaps for dimensionality reduction and data representation [J].
Belkin, M ;
Niyogi, P .
NEURAL COMPUTATION, 2003, 15 (06) :1373-1396
[7]   Forecasting Traffic Flow in Big Cities Using Modified Tucker Decomposition [J].
Bhanu, Manish ;
Priya, Shalini ;
Dandapat, Sourav Kumar ;
Chandra, Joydeep ;
Mendes-Moreira, Joao .
ADVANCED DATA MINING AND APPLICATIONS, ADMA 2018, 2018, 11323 :119-128
[8]  
Bhanu M, 2018, INT CONF COMMUN SYST, P121, DOI 10.1109/COMSNETS.2018.8328188
[9]   Deep Multi-Scale Convolutional LSTM Network for Travel Demand and Origin-Destination Predictions [J].
Chu, Kai-Fung ;
Lam, Albert Y. S. ;
Li, Victor O. K. .
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS, 2020, 21 (08) :3219-3232
[10]  
Cichocki A, 2016, FOUND TRENDS MACH LE, V9, P431, DOI [10.1561/2200000059, 10.1561/2200000067]