Multivariate Methods • Cluster Analysis • Additional Examples of Cluster Analysis
Previous
 • 
Next
Additional Examples of Cluster Analysis
Example of Self-Organizing Maps
This example uses the Iris.jmp sample data table, which includes measurements of sepal length, sepal width, petal length, and petal width for three species of irises.
1.
Select Help > Sample Data Library and open Iris.jmp.
2.
Select Analyze > Multivariate Methods > Cluster.
3.
Assign Sepal length, Sepal width, Petal length, and Petal width as Y, Column variables.
4.
Select K Means on the Options menu.
5.
Uncheck Columns Scaled Individually.
6.
Click OK.
7.
Select Self Organizing Map from the Method menu on the Control Panel.
8.
Even though we know the data consists of three species, set Optional range of clusters equal to 10.
9.
Set N Rows equal to 1 and N Columns equal to 3.
10.
Click Go.
The results are displayed in Self Organizing Map Report Window. Notice the number of clusters that gives the largest CCC is 3, which is the number of species. We can see the classification was not perfect; each cluster should represent each species, with 50 rows for each.
Self Organizing Map Report Window
11.
In the data table, select the Species column and select Rows > Color or Mark by Column.
12.
Select the Classic option under Markers.
13.
Click Go.
14.
From the SOM Grid 3 by 1 red triangle menu, select Biplot.
Biplot of Iris Self Organizing Map
We can see that all rows from Cluster 3 are correctly identified as the setosa species. The other two species, virginica and versicolor, overlap slightly and can be mistaken for each other.
15.
From the SOM Grid 3 by 1 red triangle menu, select Parallel Coord Plots.
Parallel Coordinate Plot for Iris Data
We can see from the Parallel Coordinate Plot in Parallel Coordinate Plot for Iris Data that clusters 1 and 2 (species virginica and versicolor, respectively) can be similar to each other in characteristics. These similarities can make it hard to distinguish between the species. However, the SOM did a relatively good job identifying and classifying these three species.