Cluster Interpretation: Mosaic Analysis

Background & Context

On this page, we present the Mosaic Analysis for cluster interpretation as an alternative to Most Relevant Explanations for Cluster Interpretation.

To provide further context for Most Relevant Explanations for Cluster Interpretation, we compare several other approaches that can help interpret individual Clusters:

• Setting Evidence for Cluster Interpretation: Posterior Distributions, Relationship with Target Node, Mosaic Analysis, Posterior Mean Analysis, Segment Profile Analysis, Histograms, Tornado Diagrams,
• Optimization for Cluster Interpretation: Dynamic Profile, Target Optimization Tree

More specifically, we compare all these approaches with regard to characterizing the state $\mathrm{Cluster\ 3}$ of the Cluster Node $\mathit{Factor}_0$ in the reference network.

All analyses and instructions on this page refer to this reference network, which you can download here:

MaleClusters.xbl

XBL

Mosaic Analysis for Cluster Interpretation

With the Mosaic Analysis, we can visualize the posterior probability distributions of all nodes for each Cluster. Select all nodes in the network, including the Cluster Node $\mathit{Factor}_0$. To start the Mosaic Analysis, select Menus > Analysis > Visual > Overall > Mosaic. In the Settings window, we assign $\mathit{Factor}_0$ to the vertical axis.

Upon clicking OK, BayesiaLab opens a new window showing the Mosaic Analysis.

Click to Zoom

In the above window, the row of boxes corresponding to $\mathit{Factor}_0=\mathit{Cluster\ 3}$ is highlighted with a purple frame.

In the following screenshot, we zoom in to focus on the box that represents $P(3 - \mathit{Bicep\ Girth} \gt 36 \mid \mathit{Factor}_0=\mathit{Cluster\ 3})=93.75\%$: