bnlearn_predict_wrapper: Create a predict function for use of a bnlearn Bayesian...

In johacks/bncounterfactuals: COUNTERFACTUAL AND SUFFICIENT EXPLANATIONS FOR BAYESIAN NETWORKS

Create a predict function for use of a bnlearn Bayesian network in the search algorithms.

Description

Create a predict function for use of a bnlearn Bayesian network in the search algorithms.

Usage

bnlearn_predict_wrapper(
  bn,
  node,
  method,
  threshold = NULL,
  default = NULL,
  n = 500
)

Arguments

bn	Fitted Bayesian network
node	Name of target node to be predicted
method	Same options as in parameter 'method' in predict function of bnlearn
threshold	Minimum value of posterior probability of the MLE to be the output of the algorithm. If parameter 'default' is null, the MLE will be the output even if it doesn't surpass the threshold.
default	Default predicted target value when threshold isn't exceeded by the MLE
n	Same options as in parameter 'n' in predict function of bnlearn

Value

Function to be used in algorithms such as bfs_sfx

Examples

library(bnlearn)
set.seed(40)
download.file("https://www.bnlearn.com/bnrepository/child/child.rda",
  "child.rda", "auto",
  quiet = TRUE
)
load("child.rda") # Load CHILD Bayesian Network
evidence <- data.frame(
  LVHReport = factor(x = "yes", levels = dimnames(bn$LVHreport$prob)[[1]]),
  LowerBodyO2 = factor(x = "5-12", levels = dimnames(bn$LowerBodyO2$prob)[[1]]),
  CO2Report = factor(x = "<7.5", levels = dimnames(bn$CO2Report$prob)[[1]]),
  XrayReport = factor(x = "Oligaemic", levels = dimnames(bn$XrayReport$prob)[[1]])
)
outcome <- predict(
  object = bn, node = bn$Disease$node, data = evidence,
  method = "bayes-lw"
)
expected <- factor("PAIVS", levels = levels(outcome))
predict_f <- bnlearn_predict_wrapper(bn, bn$Disease$node, "bayes-lw",
  threshold = 0.20, default = "TGA"
)
print(bfs_sfx(predict_f, evidence, outcome, expected))

johacks/bncounterfactuals documentation built on March 29, 2022, 10:13 p.m.