Dictionary (9.0)
Overview & Context
- BayesiaLab provides a convenient way for saving specific attributes of a Bayesian network, such as the properties of arcs, nodes, and states, in the form of Dictionaries.
- Dictionaries are simple text files that can be opened and edited with any text editor. For instance, this makes it easy to apply edits to many nodes at the same time.
- You can import a Dictionary again to the same network or to another network, for which you want to adopt the characteristics defined in the Dictionary.
New Feature: Structural Priors
- In BayesiaLab 9, you can now define prior knowledge for individual Arcs by assigning a so-called Structural Prior within a range of -1.0 to 1.0.
- This new attribute can be imported and exported in the form of a Dictionary.
- The syntax of a Dictionary containing Structural Priors is analogous to that of Dictionaries of other attributes.
- As an alternative to the equal sign "=" shown in the example below, you can also use space or tab.
FBS->K=0.076695684161
Condition->K=1.0
BP->K=0.076695684161
EF-TTE->FBS=0.068965584386
Condition->FBS=1.0
DM->FBS=1.0
PR->FBS=0.076695684161If you want to define a Dictionary from scratch, but don't remember the syntax, you can simply define a Structural Prior on a random Arc and then export the Dictionary. Now you have a correct code sample as a reference.
Usage
-
To export a Dictionary of the Structural Priors of your current Bayesian network, select Data > Export Dictionary > Arc > Structural Priors.
-
To import a Dictionary of Structural Priors to your current Bayesian network, select Data > Associate Dictionary > Arc > Structural Priors.
Updated Feature: Export Dictionary — Arc
-
If a subset of arcs is selected as you start the Export Dictionary process, BayesiaLab will prompt you as to whether you want All Arcs or only the Selected Arcs to be processed, i.e., included in the Dictionary file.
