19 June 2001
SPATIAL DATA MINING USING FUZZY LOGIC IN AN OBJECT-ORIENTED GEOGRAPHICAL INFORMATION DATABASE
M. Cobb
University of Southern Mississippi
Hattiesburg, MS 39406-5106
M. Chung R. Wilson K. Shaw
Naval Research Laboratory
Stennis Space Center, MS 39529-5004
F. Petry
Tulane University
New Orleans, LA 70118
Abstract
The Mapping Sciences Section of the Naval Research Laboratory, Stennis Space Center, has realized the enormous benefits of spatial data warehousing and database integration with the implementation of the Geospatial Information Database (GIDB). An object-oriented approach was used to develop an object model that could be easily expanded to include all geographic data types. With the base of object-oriented technology, standards such as Common Object Request Broker Architecture (CORBA) and Virtual Reality Modeling Language (VRML) enabled 2-dimensional as well as 3-dimensional display over the internet.
However, in the process of developing the GIDB system, the question of what to do with all the data became an inevitable question. Data exist to be used and exploited by users, but what can users do with all the data? Is the availability of so much information overwhelming to the users? The use of spatial data mining techniques to help users make sense of the wealth of data in the GIDB is the focus of this paper. After general discussions of the topic of spatial data mining, we then present a specific technique for integrating a fuzzy set model for spatial relationship determination with the object-oriented model of the GIDB.
Sponsored by the Marine Corps Warfighting Laboratory.
Presented at
the Proc. World Multiconference on Systemics, Cybernetics and Informatics, SCI/ISAS
99, July 31-August 4, 1999, Orlando, FL.
Conference Proceedings