Student research opportunities
Analysis of Spatial-temporal Data in Cities
Project Code: CECS_792
This project is available at the following levels:
Honours, Summer Scholar, Masters, PhD
Keywords:
spatial statistics, graphical models, smarter city
Supervisor:
Dr Lexing XieOutline:
Modern cities and urban life are generating abundant data traces -- some are spatial layers such as roads, land use, population density; others are spatial-temporal traces, such as weather, traffic, crime incidents, housing prices.
How do we make use of such diverse layers of information, how can they be leveraged for better decision making? What computational tool suits each task?
Goals of this project
The project goals can involve a subset or a combination of the following:
* Design and perform creative analysis to reveal novel spatial temporal patterns in urban data layers.
* Demonstrate the effectiveness of spatial temporal models in real-world applications.
* Design and develop interactive and dynamic visualization to help create insights into urban dynamics.
Requirements/Prerequisites
Knowledge of probability, statistics and lineara algebra.
Ability to program, and willingness to learn.
Background Literature
"Analytics-driven asset management", A Hampapur et al. IBM Journal of Research and Development, Jan 2011
A vision for smarter cities
http://www-03.ibm.com/innovation/us/thesmartercity/index_flash.html
A method for spatial–temporal forecasting with an application to real estate prices. Pace et al. http://dx.doi.org/10.1016/S0169-2070(99)00047-3
Petterns in Crime, P. Brantingham, 1984
Empirical macroscopic features of spatial-temporal traffic patterns at highway bottlenecks, Kerner, Phys. Rev. E 65, 046138 (2002) [http://pre.aps.org/abstract/PRE/v65/i4/e046138
