Cluster Interpretation: Dynamic Profile

Background & Context

• On this page, we present the Dynamic Profile 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 Cluster 3 of the Cluster Node $Factor_0$ in the reference network.
• All analyses and instructions on this page refer to this reference network, which you can download here:

Dynamic Profile for Cluster Interpretation

• We can use BayesiaLab’s optimization tools to work with more complex sets of evidence. One of these optimization tools is the Dynamic Profile.
• The Dynamic Profile uses a greedy search algorithm to simulate sets of evidence that maximize the probability of Cluster 3.
• It may seem counterintuitive to think of optimizing evidence to achieve membership in Cluster 3. After all, we cannot modify body measurements.
• However, we can think of those characteristics that assign a subject to Cluster 3 most quickly as prototypical traits of Cluster 3.
• To start Dynamic Profile, select Main Menu > Analysis > Target Optimization > Dynamic Profile.
• In the Dynamic Profile Settings, we need to specify that we want to maximize the probability of Cluster 3 by searching across Hard Evidence.

• Clicking OK starts the search and quickly produces a solutions table, which opens in a new window:

• The starting point is the row in the table marked A priori, which shows that $P(Factor_0=C3)=12.955$, i.e., the marginal probability of C3.
• If we evaluate the hypothesis 3 - Bicep Girth > 36, the posterior probability increases, i.e., $P(Factor_0=C3 \mid 3 - Bicep Girth > 36)=41.096$.
• In this case, the Bayes Factor is 3.172, meaning that with this hypothesis, C3 membership is 3.172 times more probable than without it.
• Setting 7 - Hip Girth ≤ 101.8, the probability of C3 membership increases further to 72.938%.
• With 4 - Navel Girth ≤ 85.2 and 6 - Forearm Girth > 29.4, we reach a C3 probability of 100%.
• As a result, we can interpret this set of evidence as a prototypical profile for a subject in C3:
  • 3 - Bicep Girth > 36
  • 7 - Hip Girth ≤ 101.8
  • 4 - Navel Girth ≤ 85.2
  • 6 - Forearm Girth > 29.4