Contents

Context

New Feature: Observable Markov Blanket

The Observable Markov Blanket is recursively defined as:

  • the observable nodes of the Markov Blanket,
  • the nodes that belong to the Observable Markov Blanket of the Not Observable nodes of the Markov Blanket.

This new feature allows to select the nodes that belong to the Observable Markov Blanket of the node on which the contextual menu has been activated. When nodes from the Markov Blanket are Not Observable, the Observable Markov Blanket of these nodes is selected instead.

These nodes can also be highlighted by using the Neighborhood Analysis function.

Example

Let's suppose we want to identify the nodes that belong to the Observable Markov Blanket of Target in the network below.

The nodes that belong to the Markov Blanket are the Parents P1 and P2 (with a green tag), the Children C1 and C2 (with a red tag), and the Spouse S1 (with a blue tag). 

The nodes P2, C1, N6, and N17 are Not Observable, as indicated by the icon .

After selecting the Observable Markov Blanket of Target, we have highlighted the selected nodes by setting them as Badges in the graph below.