liver (version 1.19)

kNN: k-Nearest Neighbour Classification

Description

kNN is used to perform k-nearest neighbour classification for test set using training set. For each row of the test set, the k nearest (based on Euclidean distance) training set vectors are found. then, the classification is done by majority vote (ties broken at random). This function provides a formula interface to the class::knn() function of R package class. In addition, it allows normalization of the given data using the scaler function.

Usage

kNN(formula, train, test, k = 1, scaler = FALSE, type = "class", l = 0, 
    use.all = TRUE, na.rm = FALSE)

Value

When type = "class" (default), a factor vector is returned, in which the doubt will be returned as NA. When type = "prob", a matrix of confidence values is returned (one column per class).

Arguments

formula

a formula, with a response but no interaction terms. For the case of data frame, it is taken as the model frame (see model.frame).

train

data frame or matrix of train set cases.

test

data frame or matrix of test set cases.

k

number of neighbours considered.

scaler

a character with options FALSE (default), "minmax", and "zscore". Option "minmax" means no transformation. This option allows the users to use normalized version of the train and test sets for the kNN aglorithm.

type

either "class" (default) for the predicted class or "prob" for model confidence values.

l

minimum vote for definite decision, otherwise doubt. (More precisely, less than k-l dissenting votes are allowed, even if k is increased by ties.)

use.all

controls handling of ties. If true, all distances equal to the kth largest are included. If false, a random selection of distances equal to the kth is chosen to use exactly k neighbours.

na.rm

a logical value indicating whether NA values in x should be stripped before the computation proceeds.

Author

Reza Mohammadi and Kevin Burke

References

Ripley, B. D. (1996) Pattern Recognition and Neural Networks. Cambridge.

Venables, W. N. and Ripley, B. D. (2002) Modern Applied Statistics with S. Fourth edition. Springer.

See Also

kNN, scaler

Examples

data(risk)

train = risk[1:100, ]
test  = risk[  101, ]

kNN(risk ~ income + age, train = train, test = test)