Spatial Data Calculator
Skip Navigation Links
Home
Point data
NN distances
All distances
Local clustering
Neighbour categories
K(t) and O(r) function
G(r), F(r), J(r) function
Spatial clusters
Thinning
Gaps
Local statistics
Interpolation
Directions
Linear data
Gradient
Line generalization
Statistics along a line
Relationships
Similarity and distance
Non-spatial clusters
Mantel test
Categorical overlay
Variogram
Autocorrelation
Local autocorrelation
Spatial correlation
Local correlation
3D autocorrelation
Non-spatial correlation
Non-spatial regression
Patterns
Random points
Suitability mapping
Frequency
Map sheet
Estonia
Probability
Map sheet
Estonia
kNN
Map sheet
Estonia
dNN
Map sheet
Estonia
SumSim
Map sheet
Estonia
Tests
Univariate sample
Difference tests
χ2 test
Classification
ROC curve
Regions of frequency
Coordinates
This function has been used 81 times
Similarity to absent or find sites (SumSim algorithm)
Source data
Input exemplar cases must be in three columns: 1 or 0 in the first column meaning find site or absence site, Lambert-Est X coordinate in the second and Y coordinate in the third column. Set direction of the axes from the radio button above the input cell. Every case must be in a separate row, column separator can be either space, tab or semicolon, decimal separator can be either point or comma. The default probability is the proportion of find sites in the learning data. Lines beginning with a character are excluded. SumSim algorithm is selecting the most similar exemplars until the total amount of similarity exceeds the given value. The SumSim algorithm is used in the software system
Constud
. See chapter 4.1.1 in the
Constud Tutorial
and an explanatory
figure
.
Area borders
E min
E max
N min
N max
Boundaries from min and max values
Estonian boundaries
Map sheet boundaries
1: 20 000 map sheet
Weights of features
Land use
Land use diversity
Land cover
10 types of soil
24 types of soil
60 types of soil
Elevation
Relative elevation
Human population density
Distance weighted human density
Total similarity searched for decisison
SD from learning cases
SD from output map
N direction first
E direction first
map background
||
||