Transcript Special Topics in Geo-Business Data Analysis

Special Topics in Geo-Business Data Analysis
Week 4 Covering Topic 7
Spatial Data Mining
Visualizing Map Surfaces
What spatial
relationships do you
see?
…do relatively high
levels of housing
Density often occur with
high levels of Value and
Age?
…how often?
…where?
…what coincidence
pattern occurs most
often?
(Berry)
Identifying Unusually High Measurements
…isolate areas with mean + 1 StDev (tail of normal curve)
(Berry)
Level Slicing
…simply multiply the two maps to identify joint coincidence
1*1=1 coincidence (any 0 results in zero)
(Berry)
Multivariate Data Space
…sum of a binary progression (1, 2 ,4 8, 16, etc.) provides
level slice solutions for many map layers
(Berry)
Linking Geographic and Data Space
Similar data patterns plot close to one another
with increasing data distance indicating less similarity
Data Distance
Pythagorean – Data Distance= (SQRT (A^2 + B^2 + C^2 …))
(Berry)
Identifying Map Similarity
…the relative data distance between the comparison point’s data pattern
and those of all other map locations form a Similarity Index
(Berry)
Examples of Map Clustering
…similarity among data patterns (data distance) is used to divide an area into “clusters”
(Berry)
Identifying Data Pattern Zones
…groups of “floating balls” in data space identify locations in a project
area that have similar data patterns– data zones
…that are 1) as different as possible between groups
and 2) as similar as possible within a group
(Berry)
Evaluating Clustering Results
…graphical and statistics procedures assess how “distinct” clusters are—
Clustering Performance
(Berry)
Exporting Map Layers
…the analysis grid provides spatially consistent formatting that is easily exported to
other software packages– each cell is a record with map values defining its fields
(Berry)
Vector to/from Raster
Pseudo Grid– each grid cell is stored
as a polygon
…pseudo grid themed for travel-time
V to R– burning the points, lines and areas into
the grid (fat, thin and split)
R to V– connecting grid centroids, sides and
edges (line smoothing)
(Berry)