parentValues: Value of parent nodes

In MoTBFs: Learning Hybrid Bayesian Networks using Mixtures of Truncated Basis Functions

Value of parent nodes

Description

This function returns a data.frame of dimension '1xn' containing the values of the 'n' parents of a 'node' of interest. Use this function if you have a random sample and an observed sample with information about the parents. The values of the parents are obtained from the evidence set unless they are not observed. In this case, the values are taken from the random sample.

Usage

parentValues(node, bn, obs, rdf)

Arguments

node	A character string that represents the node's name.
bn	A list of lists obtained from the function MoTBFs_Learning. It contains the conditional density functions of the bayesian network.
obs	A data.frame of dimension '1xm' containing an instance of the 'm' variables that belong to the evidence set.
rdf	A data.frame of dimension '1xk' containing an instance of the 'k' variables sampled from the bayesian network.

Value

A data.frame containing the values of the parents of 'node'.

Examples

## Dataset
  data("ecoli", package = "MoTBFs")
  data <- ecoli[,-c(1,9)]

## Get directed acyclic graph
  dag <- LearningHC(data)
  
## Learn bayesian network
  bn <- MoTBFs_Learning(dag, data = data, numIntervals = 4, POTENTIAL_TYPE = "MTE")
  
## Specify the evidence set
  obs <- data.frame(lip = "1", alm1 = 0.5, stringsAsFactors=FALSE)
  
## Create a random sample
  contData <- data[ ,which(lapply(data, is.numeric) == TRUE)]
  fx <- lapply(contData, univMoTBF, POTENTIAL_TYPE = "MTE")
  disData <- data[ ,which(lapply(data, is.numeric) == FALSE)]
  conSample <- lapply(fx, rMoTBF, size = 1)
  disSample <- lapply(unique(disData), sample, size = 1)
  
  rdf <- as.data.frame(list(conSample,disSample), stringsAsFactors = FALSE)
  
## Get the values of the parents of node "alm2"
  parentValues("alm2", bn, obs, rdf)

MoTBFs documentation built on Aug. 8, 2025, 7:07 p.m.