What's New:

We just released a new software package for graph partitioning and flow mapping of large spatial interaction data, such as county-to-county migration. You can download and try the software with the migration data that we share at the above link or using your own data. It involves two programs: one for graph partitioning and optimization and the other one for flow mapping with the derived regions. They are separated for the convenience of development. In the future we'll merge them into one package.

JLBS publication: Graph-based Vehicle Trajectory Analysis

Submitted by guod on 08/19/2010

It is difficult to extract meaningful patterns from massive trajectory data. One of the main challenges is to characterize, compare, and generalize trajectories to find overall patterns and trends. This research proposes a graph-based approach that converts trajectory data to a graph based representation, treats it as a complex network, and detects hidden community structures that can help characterize trajectories. [ Manuscript ]

IJGIS Publication: Local Entropy Map...

Submitted by guod on 06/27/2010

Guo, D (2010). "Local Entropy Map: A Non-Parametric Approach to Discover Spatially Varying Multiivariate Relationships". International Journal of Geographical Information Science. (in press).  Manuscript ...

IEEE TVCG Publication on Flow Mapping

Submitted by guod on 06/02/2010

Guo, D. (2009), "Flow Mapping and Multivariate Visualization of Large Spatial Interaction Data", IEEE Transactions on Visualization and Computer Graphics, vol. 15, no. 6, pp. 1041-1048. [Article Link]

AAG 2010 Special Session on Data Mining (3 sessions)

Submitted by guod on 02/27/2010

I am organizing special sessions on "Spatial Data Mining and Exploratory Data Analysis". >>>AAG Meeting

CEUS Special Issue: Spatial Data Mining - Methods and Applications

Submitted by guod on 12/12/2009

A Special Issue for journal Computers, Environment and Urban Systems, vol. 33, no. 6 (November, 2009): Spatial Data Mining–Methods and Applications, edited by Diansheng Guo and Jeremy Mennis.

more news>>>