imptree (version 0.5.1)

summary.imptree: Classification with Imprecise Probabilities

Description

Summary function for an imptree object, assesses accuracy achieved on training data and further tree properties.

Usage

# S3 method for imptree
summary(object, utility = 0.65,
  dominance = c("strong", "max"), ...)
# S3 method for summary.imptree
print(x, ...)

Arguments

object

An object of class imptree. See details.

utility

Utility for the utility based accuracy measure for a vacuous prediction result (default: 0.65).

dominance

Dominace criterion to be applied when predicting classes. This may either be "strong" (default) or "max". See details at predict.imptree.

…

Further arguments are ignored at the moment.

x

an object of class summary.imptree

Value

A named list of class summary.imptree containing the tree creation call, accuracy on the training data, meta data and supplied the utility and dominance criterion for evaluation.

call

Call to create the tree

utility

Supplied utility, or its default value

dominance

Supplied dominace criterion, or its default value

sizes

List containing the overall number and number of indeterminate predictions on training data

acc

named vector containing the accuracy measures on training data with nicer names (without size information) (see predict.imptree)

meta

named vector containing the tree's depth, number of leaves and number of nodes

The printing function returns the summary.imptree object invisibly.

Details

An existence check on the stored C++ object reference is carried out at first. If the reference is not valid the original call for "object" is printed as error.

See Also

imptree, predict.imptree, for information on a single node node_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))

## summary including prediction on training data
summary(ip)                       # default prediction
summary(ip, dominance = "max")    # different prediction parameter

# }