imptree (version 0.5.1)

node_imptree: Classification with Imprecise Probabilities

Description

Access probability information of nodes

Usage

node_imptree(x, idx = NULL)
# S3 method for node_imptree
print(x, ...)

Arguments

x

An object of class imptree or node_imptree, respectively. See details.

idx

numeric or integer vector of indices specifying the sequential node access from the root node. Numeric values are coerced to integer as by as.integer (and hence truncated towards zero). If NULL the probability information of the root node are accessed.

...

Further arguments passed to print methods

Value

An object of class node_imptree containing information on the properties of the node as a list:

probint

matrix containing the bounds of the imprecise probability distribution and the absolute observed frequencies of the classification variable within the node.

depth

The depth of the node with the tree.

splitter

The name of the variable used for splitting as character; NA if node is a leaf.

children

The number of children of the node.

traindataIdx

Vector giving the indexes of the training data contained within the node

ipmodel

List giving details about the used imprecise probability model to obatin the credal set:

iptype

used IP model: "IDM", "NPI" or "NPIapprox"

s

If iptpye == "IDM" the IDM's parameter 's', otherwise this list entry is missing

The printing function returns the node_imptree object invisibly.

Details

This function acceses the properties of a specific node of an imprecise tree. An existence check on the stored C++ object reference is carried out at first. If the reference is not valid the original call for "x" is printed as error.

See Also

imptree, for global information on the generated tree summary.imptree

Examples

# NOT RUN {
data("carEvaluation")

## create a tree with IDM (s=1) to full size
## carEvaluation, leaving the first 10 observations out
ip <- imptree(acceptance~., data = carEvaluation[-(1:10),], 
  method="IDM", method.param = list(splitmetric = "globalmax", s = 1), 
  control = list(depth = NULL, minbucket = 1))

## obtain information on the root node
node_imptree(x = ip, idx = NULL)

## obtain information on the 2nd note in the 1st level
node_imptree(x = ip, idx = c(1, 2))

## reference to an invalid index and/or level generates error
# }
# NOT RUN {
node_imptree(x = ip, idx = c(1,10))  # no 10th node on 1st level
# }
# NOT RUN {
# }