PP.classify: Predict class for the test set and calculate prediction error

In PPtree: Projection pursuit classification tree

Description

After finding tree structure, predict class for the test set and calculate prediction error.

Usage

PP.classify(test.data, true.class, Tree.result, Rule, ...) 

Arguments

test.data	the test dataset
true.class	true class of test dataset if available
Tree.result	the result of PP.Tree
Rule	split rule 1 - mean of two group means 2 - weighted mean of two group means 3 - mean of max(left group) and min(right group) 4 - weighted mean of max(left group) and min(right group)
...	...

Value

A list with components:

predict.class	predicted class
predict.error	prediction error

Author(s)

Eun-kyung Lee

References

Lee E., Cook D., and Klinke, S. (2002) Projection Pursuit indices for supervised classification

See Also

PPindex.class, PP.optimize, PP.Tree

Examples

data(iris)
n <- nrow(iris)
n.train <- round(n*0.9)
train <- sample(n, n.train)

Tree.result <- PP.Tree("LDA", iris[train,5], iris[train, 1:4])
tree.train <- PP.classify(iris[train, 1:4], iris[train, 5], Tree.result,
                          Rule=1)
tree.train
tree.test <- PP.classify(iris[-train, 1:4], iris[-train, 5],
                         Tree.result, Rule=1)
tree.test

PPtree documentation built on May 2, 2019, 4:22 a.m.