Transcript Spatial Data Mining

Overview
Shashi Shekhar
Professor, Computer Science Department
[email protected]
www.cs.umn.edu/~shekhar
Teaching:
•Csci 8705: Topics in Scientific Databases
•Csci 8701: Database Research
•Csci 5708: Database II
(spring 2003)
•Csci 4707: Database I
(Fall 2002)
Service (2002-3):
•Colloquium
•Computing Committee, Departmental Web-site Redesign
Research Area:
•Spatial Databases, Spatial Data Mining
Motivating Example
“Black Hawk Down”
Mogadishu, Somalia, 10/3/1993

Soldiers trapped by roadblocks

No alternate evacuation routes
Rescue team got lost in alleys having
no planned route to crash site

18 Army Rangers and elite Delta
Force soldiers killed, 73 wounded.

( Mark Bowden, Black Hawk Down: A Story of Modern
War )
Motivating Example
• Homeland Defense
•
Facility and Base maps
Responding to a chem-bio attack
• GIS and SDBMS needed at every step!
•
•
•
•
Gathering initial conditions
• Facility location
• Weather data from NWS
• Terrain maps (State of federal Govt.)
• Building geometry (City Govt.)
Plume simulation using supercomputers
Visualizing results – map, 3D graphics
Response planning
Weather
map,
Plume
Modeling
•
New research needs
•
Q? What happens after plume simulation, visualization?
Demographics, Transportation
( Images from www.fortune.com )
Homeland Defense: Chem-Bio Portfolio
"We packed up Morgan City residents to evacuate in the a.m. on
the day that Andrew hit coastal Louisiana, but in early afternoon
the majority came back home. The traffic was so bad that
they couldn't get through Lafayette."
- Morgan City, Louisiana Mayor Tim Mott
( http://i49south.com/hurricane.htm )
( National Weather Services)
Hurrican Andrew, 1992
 Traffic congestions on all highways
 Great confusions and chaos
( www.washingtonpost.com)
Spatial Database Research at U of M
•
Spatial Data poses new challenge for Computer Science
–
–
–
–
Parallel formulations for terrain visualization
Efficient storage methods, e.g. CCAM
Scalable routing algorithms for very large maps
Spatial Data Mining
Nest locations
Vegetation
durability
Distance to open water
Water depth
Spatial Data Mining
• Co-locations
• Spatial Outliers
•
A.
Spatial Data Mining – Tele connection Patterns!