The clustertree object is the computational object for belief propagation.
A clustertree object can be obtained by the ElimTreeInitialize,
and it contains necessary information for belief propagation, such as
the topological inforamtion of Bayesian network and cluster tree, as well as the local models. This
object comprises the following elements,
nodesA list of the node or variable names.
node.classA named vector of logical values indicating whether each node is continuous or discrete.
dagA graphNEL object of the Bayesian network structure.
tree.graphA graphNEL object of the semi-elimination tree structure.
cluster.treeA list storing the cluster members and corresponding local models as linear predictor
potentials, conditional probability tables, and joint distribution tables.
assignmentA list indicating the cluster assignment of the discrete factors.
propagatedA logical value indicating whether the discrete compartment has been propagated.
absorbed.variablesA vector of characters indicating variables observed with hard evidence.
absorbed.valuesA list indicating the values of the variables observed with hard evidence.
absorbed.soft.variablesA vector of characters indicating variables observed with soft or
likelihood evidence.
absorbed.soft.valuesA list of the likelihoods of the soft or likelihood evidence.