Multi-Level Clustering and its Visualization for Exploratory Spatial Analysis
File version
Author(s)
Lee, I
Griffith University Author(s)
Primary Supervisor
Other Supervisors
Editor(s)
P. Bergougnoux; J.R. Herring; S. Shekhar
Date
Size
File type(s)
Location
License
Abstract
Exploratory spatial analysis is increasingly necessary as larger spatial data is managed in electro-magnetic media. We propose an exploratory method that reveals a robust clustering hierarchy from 2-D point data. Our approach uses the Delaunay diagram to incorporate spatial proximity. It does not require prior knowledge about the data set, nor does it require preconditions. Multi-level clusters are successfully discovered by this new method in only O(nlogn) time, where n is the size of the data set. The efficiency of our method allows us to construct and display a new type of tree graph that facilitates understanding of the complex hierarchy of clusters. We show that clustering methods adopting a raster-like or vector-like representation of proximity are not appropriate for spatial clustering. We conduct an experimental evaluation with synthetic data sets as well as real data sets to illustrate the robustness of our method.
Journal Title
GeoInformatica
Conference Title
Book Title
Edition
Volume
6
Issue
2
Thesis Type
Degree Program
School
Publisher link
Patent number
Funder(s)
Grant identifier(s)
Rights Statement
Rights Statement
Item Access Status
Note
Access the data
Related item(s)
Subject
Physical geography and environmental geoscience
Other earth sciences
Geomatic engineering