The K-means add-on enables you to perform K-Means clustering on your data within the Sisense Web Application. K-means clusters are partitioned into statistically significant groups according to measures you define by the k-means method.

The K-means add-on requires that you have an R server installed and it is configured to work with Sisense. For more information, see Connecting Sisense to your R Server.

kmeans

Through the K-Means add-on, you can identify distinct groups in your data based on how close they are to each other.

To install the K-means add-on:

  1. Download the attachment and unzip the contents into your C:\Program Files\Sisense\PrismWeb\plugins\ folder. If the plugins folder doesn’t exist, just create it. After those files have been unzipped there, you may also have to restart the web server.
  2. In the Sisense Web Application, create a scatter chart by selecting New Widget > Advanced ConfigurationScatter Chart.
  3. Define the relevant measures for your scatter chart.
  4. From the widget menu, select K-means.
    kmeans
    The Clustering window is displayed.
    clustering
  5. In the Clustering window, add up to four measures. Measures represent numeric matrices of data, or an object that can be coerced to such a matrix (such as a numeric vector or a data frame with all numeric columns).
  6. Select the number of clusters. For each cluster, Sisense assigns a color to represent the cluster as a statistically significant group within the chart.
  7. Click Apply, then click Apply in the widget editor.  The k-means clustering widget is displayed in the dashboard.