bayesia logo
BayesiaLab
Edit Reference States

Edit Reference States

Context

  • A Reference State is a node property that has a similar purpose as the Target State of a Target Node. Given that there can be only a single Target Node in a Bayesian network, the Reference State can extend this concept to all nodes.
  • The Reference State indicates which particular Node State an analysis should focus on.
  • The Reference State can be used in the context of the following analysis functions:
  • A Reference State can be defined manually using the Node Editor, or by using a Dictionary.
  • If one or more Reference States are defined, the Reference State icon is shown in the Status Bar.
  • Additionally, nodes that contain a Reference State are marked with the Reference State tag.

Usage

  • There are several ways to manage Reference States in BayesiaLab:
    • Click the ​icon to open the Reference State Editor.
    • Open the Node Editor and go to the Reference State tab.
  • Right-clicking the icon removes all the Reference States.

Dictionary Usage

  • You can define Reference States interactively node by node with the help of the Node Editor.
3408262
  • Alternatively, you can import and export Reference States with a dictionary. The dictionary syntax is the same as the one used for a Filtered State dictionary, i.e.:

    [ <NodeName> "." | <ClassName> "." ] <StateName>
  • Formal Grammar of the Reference State Dictionary:

    • If a node name is used as a prefix, a Reference State will be set only for the state of this node.
    • If a class name is used as a prefix, a Reference State will be set for the nodes of the class that have this state.
    • If no prefix is used, a Reference State will be set for all the nodes that have the specified state.
    • A state can be present only once; otherwise, the last occurrence is chosen.

Tip

  • Should you forget the syntax, simply use the Node Editor on one or two nodes and then export the dictionary. You can then open the dictionary file with a text editor and see examples of the syntax.

Copyright © 2025 Bayesia S.A.S., Bayesia USA, LLC, and Bayesia Singapore Pte. Ltd. All Rights Reserved.