R/witness.R

In bnRep: A Repository of Bayesian Networks from the Academic Literature

#' witness Bayesian Network
#'
#' Measuring coherence with Bayesian networks.
#'
#' @usage NULL
#'
#' @format
#' A discrete Bayesian modelling a situation where equally reliable witnesses try to identify a criminal. Probabilities were given within the referenced paper. The vertices are:
#' \describe{
#'   \item{W1SteveDidIt}{Witness 1 report: Steve did it (True, False);}
#'   \item{W2SteveDidIt}{Witness 2 report: Steve did it (True, False);}
#'   \item{W3SteveMartinOrDavidDidIt}{Witness 3 report: Steve, Martin, or David did it (True, False);}
#'   \item{W4SteveJohnOrJamesDidIt}{Witness 4 report: Steve, John, or James did it (True, False);}
#'   \item{W5SteveJohnOrPeterDidIt}{Witness 5 report: Steve, John, or Peter did it (True, False);}
#'   \item{WhoCommittedTheDeed}{Who is the criminal (Steve, Martin, David, John, James, Peter);}
#' }
#'
#' @return An object of class \code{bn.fit}. Refer to the documentation of \code{bnlearn} for details.
#' @keywords NaiveBayes
#' @importClassesFrom bnlearn bn.fit
#' @references Kowalewska, A., & Urbaniak, R. (2023). Measuring coherence with Bayesian networks. Artificial Intelligence and Law, 31(2), 369-395.
"witness"