Yuandong Wang
Abstract
Ride-hailing applications have been offering convenient ride services for people in need. However, such applications still suffer from the issue of supply-demand disequilibrium, which is a typical problem for traditional taxi services. With effective predictions on passenger demands, we can alleviate the disequilibrium by pre-dispatching, dynamic pricing or avoiding dispatching cars to zero-demand areas. Existing studies of demand predictions mainly utilize limited data sources, trajectory data, or orders of ride services or both of them, which also lacks a multi-perspective consideration. In this article, we present a unified framework with a new combined model and a road-network-based spatial partition to leverage multi-source data and model the passenger demands from temporal, spatial, and zero-demand-area perspectives. In addition, our framework realizes offline training and online predicting, which can satisfy the real-time requirement more easily. We analyze and evaluate the performance of our combined model using the actual operational data from UCAR. The experimental results indicate that our model outperforms baselines on both Mean Absolute Error and Root Mean Square Error on average.
Supplemental Material
Available for Download
Supplemental movie and image files for, A Unified Framework with Multi-source Data for Predicting Passenger Demands of Ride Services
- 2016. Scalable and Flexible Gradient Boosting. Retrieved from https://xgboost.readthedocs.io/en/latest/.Google Scholar
- 2017. BIRCH. Retrieved from http://scikit-learn.org/stable/modules/clustering.html.Google Scholar
- 2017. Didi Chuxing. Retrieved from https://en.wikipedia.org/wiki/Didi_Chuxing.Google Scholar
- 2017. Uber. Retrieved from https://en.wikipedia.org/wiki/Uber_(company).Google Scholar
- 2017. UCAR. Retrieved from https://www.crunchbase.com/organization/ucar.Google Scholar
- G. E. P. Box and G. M. Jenkins. 1971. Time series analysis, forecasting and control [J]. Journal of the American Statistical Association 134, 3 (1971).Google Scholar
- Ye Ding, Siyuan Liu, Jiansu Pu, and Lionel M. Ni. 2013. HUNTS: A trajectory recommendation system for effective and efficient hunting of taxi passengers. In 14th International Conference on Mobile Data Management. 107--116.Google Scholar
- T. Hastie, J. Friedman, and R. Tibshirani. 2010. The elements of statistical learning [J]. Technometrics 45, 3 (2010), 267--268.Google Scholar
- Hector Gonzalez, Jiawei Han, Xiaolei Li, Margaret Myslinska, and John Paul Sondag. 2007. Adaptive fastest path computation on a road network: A traffic mining approach. In 33rd International Conference on Very Large Data Bases. 794--805.Google Scholar
- S. Hochreiter and J. Schmidhuber. 1997. Long short-term memory. Neural Computation 9, 8 (1997), 1735--1780.Google ScholarDigital Library
- Yan Huang and Jason W. Powell. 2012. Detecting regions of disequilibrium in taxi services under uncertainty. In 20th International Conference on Advances in Geographic Information Systems. 139--148.Google Scholar
- Ren-Hung Hwang, Yu-Ling Hsueh, and Yu-Ting Chen. 2015. An effective taxi recommender system based on a spatio-temporal factor analysis model [J]. Information Sciences 314 (2015), 28--40.Google ScholarDigital Library
- Wenbo Jiang, Tianyu Wo, Mingming Zhang, Renyu Yang, and Jie Xu. 2015. A method for private car transportation dispatching based on a passenger demand model. In 2nd International Conference on Internet of Vehicles - Safe and Intelligent Mobility.Google ScholarDigital Library
- Yann LeCun, Yoshua Bengio, and Geoffrey Hinton. 2015. Deep learning. Nature 521, 7553 (2015), 436.Google Scholar
- Bin Li, Daqing Zhang, Lin Sun, Chao Chen, Shijian Li, Guande Qi, and Qiang Yang. 2011. Hunting or waiting? discovering passenger-finding strategies from a large-scale real-world taxi dataset. In 2011 IEEE International Conference on Pervasive Computing and Communications Workshops. 63--68.Google ScholarCross Ref
- Xiaolong Li, Gang Pan, Zhaohui Wu, Guande Qi, Shijian Li, Daqing Zhang, Wangsheng Zhang, and Zonghui Wang. 2012. Prediction of urban human mobility using large-scale taxi traces and its applications. Frontiers of Computer Science 6, 1 (2012), 111--121.Google ScholarCross Ref
- Yangdong Liu, Ye Tian, Bo Yuan, Chang Wu, Weishuo Qian, and Wei Mao. 2014. Providing useful information for passengers based on TTF model. In 2014 IEEE International Conference on Internet of Things (iThings), and IEEE Green Computing and Communications (GreenCom) and IEEE Cyber, Physical and Social Computing (CPSCom). 39--44.Google ScholarDigital Library
- Ye Liu, Yu Zheng, Yuxuan Liang, Shuming Liu, and David S. Rosenblum. 2016. Urban water quality prediction based on multi-task multi-view learning. In 25th International Joint Conference on Artificial Intelligence. 2576--2581.Google Scholar
- Qi Luo, Junming Zhang, Zhihan Liu, and Jinglin Li. 2014. On discovering regional taxi service disequilibrium with geographical collaborative filtering. In International Conference on Informative and Cybernetics for Computational Social Systems. 51--56.Google Scholar
- Luis Moreira-Matias, João Gama, Michel Ferreira, and Luís Damas. 2012. A predictive model for the passenger demand on a taxi network. In 15th International IEEE Conference on Intelligent Transportation Systems. 1014--1019.Google ScholarCross Ref
- Luís Moreira-Matias, João Gama, Michel Ferreira, João Mendes-Moreira, and Luís Damas. 2012. Online predictive model for taxi services. In 11th international conference on Advances in Intelligent Data Analysis. 230--240.Google ScholarDigital Library
- Luis Moreira-Matias, Joao Gama, Michel Ferreira, Joao Mendes-Moreira, and Luis Damas. 2013. On predicting the taxi-passenger demand: A real-time approach. In Portuguese Conference on Artificial Intelligence. 54--65.Google ScholarCross Ref
- Luis Moreira-Matias, Joao Gama, Michel Ferreira, João Mendes-Moreira, and Luis Damas. 2013. Predicting taxi--passenger demand using streaming data. IEEE Transactions on Intelligent Transportation Systems 14, 3 (2013), 1393--1402.Google ScholarDigital Library
- Luís Moreira-Matias, João Gama, Michel Ferreira, João Mendes-Moreira, and Luis Damas. 2016. Time-evolving O-D matrix estimation using high-speed GPS data streams [J]. Expert Systems with Applications 44, C (2016), 275--288.Google Scholar
- Santi Phithakkitnukoon, Marco Veloso, Carlos Bento, Assaf Biderman, and Carlo Ratti. 2010. Taxi-aware map: Identifying and predicting vacant taxis in the city. In International Joint Conference on Ambient Intelligence (2010). 86--95.Google ScholarCross Ref
- Guande Qi, Gang Pan, Shijian Li, Zhaohui Wu, Daqing Zhang, Lin Sun, and Laurence Tianruo Yang. 2013. How long a passenger waits for a vacant taxi--large-scale taxi trace mining for smart cities. In IEEE International Conference on Green Computing and Communications and IEEE Internet of Things and IEEE Cyber, Physical and Social Computing. 1029--1036.Google ScholarDigital Library
- Haochen Tang, Michael Kerber, Qixing Huang, and Leonidas Guibas. 2013. Locating lucrative passengers for taxicab drivers. In 21st ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems. 504--507.Google ScholarDigital Library
- Robert Tibshirani. 1996. Regression shrinkage and selection via the lasso. Journal of the Royal Statistical Society Series B (Methodological) 58, 1 (1996), 267--288.Google ScholarCross Ref
- Hua Wei, Yuandong Wang, Tianyu Wo, Yaxiao Liu, and Jie Xu. 2016. ZEST: A hybrid model on predicting passenger demand for chauffeured car service. In 25th ACM International on Conference on Information and Knowledge Management. 2203--2208.Google ScholarDigital Library
- Jing Yuan, Yu Zheng, Liuhang Zhang, Xing Xie, and Guangzhong Sun. 2011. Where to find my next passenger. In 13th International Conference on Ubiquitous Computing. 109--118.Google ScholarDigital Library
- Nicholas Jing Yuan, Yu Zheng, and Xing Xie. 2012. Segmentation of Urban Areas Using Road Networks [J]. Microsoft Technical Report.Google Scholar
- Desheng Zhang, Tian He, Shan Lin, Sirajum Munir, and John Stankovic. 2014. Dmodel: Online taxicab demand model from big sensor data in a roving sensor network. In IEEE International Congress on Big Data. 152--159.Google ScholarDigital Library
- Desheng Zhang, Tian He, Shan Lin, Sirajum Munir, and John A. Stankovic. 2015. Online cruising mile reduction in large-scale taxicab networks. IEEE Transactions on Parallel and Distributed Systems 26, 11 (2015), 3122--3135.Google ScholarDigital Library
- Junbo Zhang, Yu Zheng, and Dekang Qi. 2017. Deep spatio-temporal residual networks for citywide crowd flows prediction. In 31st AAAI Conference on Artificial Intelligence. 1655--1661.Google Scholar
- Kai Zhang, Zhiyong Feng, Shizhan Chen, Keman Huang, and Guiling Wang. 2016. A framework for passengers demand prediction and recommendation. In IEEE International Conference on Services Computing. 340--347.Google ScholarCross Ref
- Kai Zhao, Denis Khryashchev, Juliana Freire, Claudio Silva, and Huy Vo. 2016. Predicting taxi demand at high spatial resolution: Approaching the limit of predictability. In IEEE International Conference on Big Data. 833--842.Google ScholarCross Ref
- Xudong Zheng, Xiao Liang, and Ke Xu. 2012. Where to wait for a taxi? In ACM SIGKDD International Workshop on Urban Computing. 149--156.Google ScholarDigital Library
- Yu Zheng, Furui Liu, and Hsun Ping Hsieh. 2013. U-Air: When urban air quality inference meets big data. In 19th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 1436--1444.Google ScholarDigital Library
- Yu Zheng, Xiuwen Yi, Ming Li, Ruiyuan Li, Zhangqing Shan, Eric Chang, and Tianrui Li. 2015. Forecasting fine-grained air quality based on big data. In 21st ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2267--2276.Google ScholarDigital Library
- Qi Zhou, Junming Zhang, Jinglin Li, and Shangguang Wang. 2014. Discovering regional taxicab demand based on distribution modeling from trajectory data. In IEEE 17th International Conference on Computational Science and Engineering. 1605--1610.Google ScholarDigital Library
Index Terms
- A Unified Framework with Multi-source Data for Predicting Passenger Demands of Ride Services
Recommendations
Passenger Trip Planning using Ride-Sharing Services
CHI '18: Proceedings of the 2018 CHI Conference on Human Factors in Computing SystemsRide-sharing can potentially address transportation challenges such as traffic congestion and air pollution by letting drivers share their cars unused capacity with a number of passengers. However, even though multiple ride-sharing services exist and ...
Predicting Multi-step Citywide Passenger Demands Using Attention-based Neural Networks
WSDM '18: Proceedings of the Eleventh ACM International Conference on Web Search and Data MiningPredicting passenger pickup/dropoff demands based on historical mobility trips has been of great importance towards better vehicle distribution for the emerging mobility-on-demand (MOD) services. Prior works focused on predicting next-step passenger ...
Predicting the Dynamic Demand of Bike-Sharing System in Chicago with Divvy Operation Data: A Data-Driven approach for bike-sharing demand forecasting
ICEEG '21: Proceedings of the 5th International Conference on E-Commerce, E-Business and E-GovernmentDuring the recent decade, bike-sharing programs have increasingly become an irreplaceable part of the public transportation system. However, at the same time when people start to recognize its convenience, the ever-growing demand of the sharing bikes ...
Reviews
Amos O Olagunju
Ride-sharing service customers look for and deserve fair fares. However, with the use of the Internet to access competing fares when booking shared rides, how should ride-sharing providers forecast passenger demands in order to remain competitive Wang et al. offer a new framework for predicting ride-sharing demands originating from different data sources. The authors concisely review literature on fair prediction algorithms. The existing algorithms-ride-finding, passenger-finding, and global dispatching-are deficient because they (a) undersupply rides that satisfy the timely demands of passengers, and (b) oversupply scheduled rides that increase the delay times for drivers to pick up new passengers. Consequently, the authors present a framework for investigating "the bias between the trajectory data and the real ground truth." The authors offer unique contributions for effectively studying the ride-sharing patterns: (1) a parameter for differentiating ride time rates in alternative areas, and (2) a machine learning model, predicated on a variety of global positioning system (GPS) meteorological datasets, to effectively target fare-sharing riders and drivers based on big data analyses. Numerous experiments were performed with real-life datasets to ascertain the effectiveness of the proposed model. The experimental results compare favorably with the well-known results of statistical and machine learning algorithms in the literature. Even though the experimental dataset is limited to one region, there is no doubt that researchers can replicate the experiments given the continued challenges and research opportunities related to the Internet of Things (IoT).
Access critical reviews of Computing literature here
Become a reviewer for Computing Reviews.
Comments