A Multi-Task Matrix Factorized Graph Neural Network for Co-Prediction of Zone-Based and OD-Based Ride-Hailing Demand

被引:50
作者
Feng, Siyuan [1 ]
Ke, Jintao [2 ]
Yang, Hai [1 ]
Ye, Jieping [3 ]
机构
[1] Hong Kong Univ Sci & Technol, Dept Civil & Environm Engn, Hong Kong, Peoples R China
[2] Hong Kong Polytech Univ, Dept Logist & Maritime Studies, Hong Kong, Peoples R China
[3] Univ Michigan, Dept Computat Med & Bioinformat, Ann Arbor, MI 48109 USA
关键词
Predictive models; Task analysis; Decoding; Correlation; Transportation; Data models; Semantics; Ride-hailing; OD-based prediction; mixture-model graph convolutional network; matrix factorization; deep multi-task learning;
D O I
10.1109/TITS.2021.3056415
中图分类号
TU [建筑科学];
学科分类号
0813 ;
摘要
Ride-hailing service has witnessed a dramatic growth over the past decade but meanwhile raised various challenging issues, one of which is how to provide a timely and accurate short-term prediction of supply and demand. While the predictions for zone-based demand have been extensively studied, much less efforts have been paid to the predictions for origin-destination (OD) based demand (namely, demand originating from one zone to another). However, OD-based demand prediction is even more important and worth further explorations, since it provides more elaborate trip information in the near future as reference for fine-grained operations, such as the routing and matching of shared ride-hailing services that pick up and drop off two or more passengers in each ride. Simultaneous prediction of both zone-based and OD-based demand can be an interesting and practical problem for the ride-hailing platforms. To address the issue, we propose a multi-task matrix factorized graph neural network (MT-MF-GCN), which consists of two major components: (1) a GCN (graph convolutional network) basic module that captures the spatial correlations among zones via mixture-model graph convolutional (MGC) network, and (2) a matrix factorization module for multi-task predictions of zone-based and OD-based demand. By evaluations on the real-world on-demand data in Manhattan and Haikou, we show that the proposed model outperforms the state-of-the-art baseline methods in both zone- and OD-based predictions.
引用
收藏
页码:5704 / 5716
页数:13
相关论文
共 41 条
[1]  
Ahmed M. S., 1979, Transp. Res. Rec.
[2]   Dynamic data-driven local traffic state estimation and prediction [J].
Antoniou, Constantinos ;
Koutsopoulos, Haris N. ;
Yannis, George .
TRANSPORTATION RESEARCH PART C-EMERGING TECHNOLOGIES, 2013, 34 :89-107
[3]   Neural-Network-Based Models for Short-Term Traffic Flow Forecasting Using a Hybrid Exponential Smoothing and Levenberg-Marquardt Algorithm [J].
Chan, Kit Yan ;
Dillon, Tharam S. ;
Singh, Jaipal ;
Chang, Elizabeth .
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS, 2012, 13 (02) :644-654
[4]   A Context-Aware Nonnegative Matrix Factorization Framework for Traffic Accident Risk Estimation via Heterogeneous Data [J].
Chen, Quanjun ;
Song, Xuan ;
Fan, Zipei ;
Xia, Tianqi ;
Yamada, Harutoshi ;
Shibasaki, Ryosuke .
IEEE 1ST CONFERENCE ON MULTIMEDIA INFORMATION PROCESSING AND RETRIEVAL (MIPR 2018), 2018, :346-351
[5]  
Defferrard M, 2016, ADV NEUR IN, V29
[6]   A bayesian dynamic linear model approach for real-time short-term freeway travel time prediction [J].
Fei, Xiang ;
Lu, Chung-Cheng ;
Liu, Ke .
TRANSPORTATION RESEARCH PART C-EMERGING TECHNOLOGIES, 2011, 19 (06) :1306-1318
[7]   Improving molecular cancer class discovery through sparse non-negative matrix factorization [J].
Gao, Y ;
Church, G .
BIOINFORMATICS, 2005, 21 (21) :3970-3975
[8]  
Geng X., 2019, ARXIV190511395
[9]  
Geng X, 2019, AAAI CONF ARTIF INTE, P3656
[10]  
Guillamet D, 2002, INT C PATT RECOG, P116, DOI 10.1109/ICPR.2002.1048251