Junbo Zhang
Yu Zheng
ABSTRACT
Advances in location-acquisition and wireless communication technologies have led to wider availability of spatio-temporal (ST) data, which has unique spatial properties (i.e. geographical hierarchy and distance) and temporal properties (i.e. closeness, period and trend). In this paper, we propose a <u>Deep</u>-learning-based prediction model for <u>S</u>patio-<u>T</u>emporal data (DeepST). We leverage ST domain knowledge to design the architecture of DeepST, which is comprised of two components: spatio-temporal and global. The spatio-temporal component employs the framework of convolutional neural networks to simultaneously model spatial near and distant dependencies, and temporal closeness, period and trend. The global component is used to capture global factors, such as day of the week, weekday or weekend. Using DeepST, we build a real-time crowd flow forecasting system called UrbanFlow1. Experiment results on diverse ST datasets verify DeepST's ability to capture ST data's spatio-temporal properties, showing the advantages of DeepST beyond four baseline methods.
- M. X. Hoang, Y. Zheng, and A. K. Singh. Forecasting citywide crowd flows based on big data. ACM SIGSPATIAL 2016, October 2016.Google Scholar
- A. Krizhevsky, I. Sutskever, and G. E. Hinton. Imagenet classification with deep convolutional neural networks. In Advances in neural information processing systems, pages 1097--1105, 2012.Google ScholarDigital Library
- Y. Li, Y. Zheng, H. Zhang, and L. Chen. Traffic prediction in a bike-sharing system. In Proceedings of the 23rd SIGSPATIAL International Conference on Advances in Geographic Information Systems, page 33. ACM, 2015. Google ScholarDigital Library
- Y. Zheng, L. Capra, O. Wolfson, and H. Yang. Urban computing: concepts, methodologies, and applications. ACM Transactions on Intelligent Systems and Technology (TIST), 5(3):38, 2014. Google ScholarDigital Library
- Y. Zheng, F. Liu, and H.-P. Hsieh. U-air: When urban air quality inference meets big data. In Proceedings of the 19th ACM SIGKDD international conference on Knowledge discovery and data mining, pages 1436--1444. ACM, 2013. Google ScholarDigital Library
Recommendations
A spatio-temporal decomposition based deep neural network for time series forecasting
AbstractSpatio-temporal problems arise in a broad range of applications, such as climate science and transportation systems. These problems are challenging because of unique spatial, short-term and long-term patterns, as well as the curse of ...
Highlights- A deep neural network is proposed for the short-term spatio-temporal forecasting.
To approach cylindrical coordinates to represent multivariable spatio-temporal data
ICCCI'12: Proceedings of the 4th international conference on Computational Collective Intelligence: technologies and applications - Volume Part IIData representing a moving object include the data of time, position, and attributes. The data of positions and attributes of a moving object, which change over time may be recorded asynchronously because of the difference of sampling methods. ...
Spatio-temporal access methods: a survey (2010 - 2017)
The volume of spatio-temporal data is growing at a rapid pace due to advances in location-aware devices, e.g., smartphones, and the popularity of location-based services, e.g., navigation services. A number of spatio-temporal access methods have been ...
Comments