PlotCGBN: Plot the Bayesian network

In BayesNetBP: Bayesian Network Belief Propagation

Plot the Bayesian network

Description

Plot and compare two Bayesian networks with different evidence(s) absorbed and propagated.

Usage

PlotCGBN(
  tree.1,
  tree.2,
  fontsize = NULL,
  pbar = FALSE,
  plotting = TRUE,
  epsilon = 10^-6
)

Arguments

tree.1	a ClusterTree
tree.2	a ClusterTree
fontsize	font size for the node labels
pbar	logical(1) whether to show progress bar
plotting	logical(1) whether to output plot
epsilon	numeric(1) the KL divergence is undefined if certain states of a discrete variable have probabilities of 0. In this case, a small positive number epsilon is assigned as their probabilities for calculating the divergence. The probabilities of other states are shrunked proportionally to ensure they sum up to 1.

Details

Network visualization of the node-specific differences between Bayesian Networks with the same topology, but evidence that has been absorbed and propagated. The change of marginal distribution of each node is measured by signed and symmetric Kullback-Leibler divergence. The sign indicates the direction of change, with tree.1 considered as the baseline. The magnitude of the change is reflected by the value. Nodes that are white are d-separated from the evidence. This function requires Rgraphviz package.

Value

a plot of Bayesian network

a vector of signed symmetric Kullback-Leibler divergence

Author(s)

Han Yu

References

Cowell, R. G. (2005). Local propagation in conditional Gaussian Bayesian networks. Journal of Machine Learning Research, 6(Sep), 1517-1550.

Yu H, Moharil J, Blair RH (2020). BayesNetBP: An R Package for Probabilistic Reasoning in Bayesian Networks. Journal of Statistical Software, 94(3), 1-31. <doi:10.18637/jss.v094.i03>.

Examples

## Not run: 
library("Rgraphviz")
data(toytree)
tree.post <- AbsorbEvidence(toytree, c("Nr1i3"), list(1))
PlotCGBN(tree.1=toytree, tree.2=tree.post)

## End(Not run)

BayesNetBP documentation built on May 9, 2022, 1:05 a.m.