Documentation and SupportV5 User Manual Using the R ToolHow do I perform a cluster analysis?

How do I perform a cluster analysis?

1. From the Foundation R Integration Tool select Clustering from the sidebar.

From the TMMData R Integration Tool select Clustering from the sidebar.

2. Choose your cluster method, and optionally, select a column to exclude from the analysis.

Choose your cluster method, and optionally, select a column to exclude from the analysis.

2.1. A column that you choose to exclude from the clustering will still appear in the final data frame.  This is a diagnostic tool to examine a categorization that already exists within the data.

A column that you choose to exclude from the clustering will still appear in the final data frame.  This is a diagnostic tool to examine a categorization that already exists within the data.

3. Select Cluster

3.1. K-Means and K-Modes Cluster

K-Means and K-Modes clustering return your original data frame with a new column attached which contains the result of the analysis.  If you excluded a column from the analysis it will appear here as well, but it will not have been used to calculate the clusters.  This table is what will be saved back to the Foundation.

3.2. Hierarchical

Foundation utilizes components from HighSoft's HighCharts.

The colored Dendrogram created by the Hieararchical algorithm shows the shape of your clusters based on the similarity of rows.  When saved back to the Foundation it will provide a table with the original data and the clusters.

3.3. Principal Component Analysis

Foundation utilizes components from HighSoft's HighCharts.

Principal Component Analysis provides a chart of the variances explained by each principal component as well as a table with loadings for each principal component.  The table is what is saved back to the Foundation.

4. Saving the results of the analysis

Saving the results of the analysis

After clicking on Save under one of your analyses enter the name for your table and click on "Save."  The system will load the data into the auto importer for use in the Product.