Child pages
  • Prior Samples (9.0)



Data | Prior Samples

The generation of Prior Samples is a way to define Dirichlet Priors. The expert prior knowledge is represented by a fully-specified Bayesian network. Basically, it can be considered as sampling particles from the joint probability distribution encoded by the current network and saving these particles in a Virtual Data Set. The actual data set and the virtual one are then used for machine learning the structure and estimating the marginal and conditional probability distributions. The number of particles defines the weight of the expert knowledge. 


Since version 5.4, generating prior samples has been done via the Learning menu. This feature has been renamed Generate Prior Samples in version 7.0.

Renamed Feature: Prior Samples

Virtual Data Set is now named Prior Samples.

Updated Feature: Generate

As of version 9.0, Generate Prior Samples has been moved from the Learning menu to the Data menu.

Upon generating the virtual dataset, the icon  appears in the lower right corner of the Graph Window. Hovering over this icon shows you the number of prior samples. You can delete the virtual data set by left-clicking . Right-clicking  brings up a view of the Bayesian network structure that has been used for representing the prior knowledge.