/
Clustering

Clustering

Clustering uses machine learning to automatically group similar data points based on multiple dimensions and measures. It identifies natural patterns and segments without predefined categories. Tableau uses the k-means algorithm, automatically determining the optimal number of clusters. Clustering reveals hidden patterns, customer segments, or product groupings for targeted analysis.

To find clusters, perform the following steps.

  1. Connect Tableau with the Kyvos semantic model and create a scatter plot like the one below.

    image-20251009-142441.png

  2. From the Analytics tab, click Analysis > Cluster > and then drag into the viz.

    image-20251009-142543.png

  3. Tableau automatically creates clusters shown in Color. Users can configure the number of clusters they want in Viz.

    image-20251009-142616.png

  4. A Clusters field appears in the Data pane. Right-click Clusters > Describe Clusters to see characteristics.

    image-20251009-142728.png

     

    image-20251009-142820.png

  5. To adjust the number of clusters, right-click the field > Edit Clusters and change the variables used for clustering in the dialog.

    image-20251009-142927.png

  6. View the values based on the configured cluster.

    image-20251009-143030.png



     

     

 

Copyright Kyvos, Inc. 2025. All rights reserved.