getMutualInformation: Estimate mutual information

In mickash/Adaptive-Bayesian-Networks: Adaptive Bayesian Networks.

Description

Estimate mutual information on network given evidence. Note the mutual information of a variable with itself is the entropy of that variable (not necessarily 1).

Usage

getMutualInformation(net, evidence, nodes, envNodes, weights = rep(1,
  length(nodes)), num = 1000, algorithm = "IS", algoSpecific = "nat",
  debug = FALSE)

Arguments

net	the network
evidence	vector of integers specifying known values of nodes. This can include NA values for unknown variables.
nodes	The nodes to estimate informative environment variables for.
envNodes	The environment variables.
weights	A weight vector to combine results for weighted overall information.
num	integer specifying the number of samples to used in inference
algorithm	A string specifying a valid sampling algorithm.
algoSpecific	Required settings for sampling algorithm.
debug	For internal use only.

Value

matrix with each row giving mutual information of node with environment variables. An additional row exists for the weighted combined information if weights is not NULL.

mickash/Adaptive-Bayesian-Networks documentation built on Nov. 14, 2019, 12:14 a.m.