lau_knn

USE:

[class]=lau_knn(k,X,y,x)

DESCRIPTION:

Implements a k-nearest neighbor algorithm. This is a simple classification rule which assigns for a new point x the class in which are found most of the k nearest neighbors of x in the training data set.

ARGUMENTS:

X must be an mxp matrix (p>=1) and y an m rows vector. y corresponds to the class variable while X with the feature vector. Values of y must be integers from 1 to Nc (number of classes).

k is the number of neighbors on which decision is taken. 

x is a matrix of size (nxp) of new samples which needs to be classified. n>=1.

VALUES:

The function returns class which is a column vector indicating the class in which the elements of x matrix are classified.

EXAMPLES:

//load the iris data set. There are 3 classes and 50 samples in each

> irisdata

//use half of the iris data set as prototypes, for a 5-nearest neighbor, to classify the remaining half

>class=lau_knn(5,X(1:2:end,:),y(1:2:end),X(2:2:end,:))

//computes in how many of the 75 cases the classification was successful

> sum(class==y(2:2:end))

COMMENTS:

Created by Laurentiu Adi Tarca. Revised 10.07.2003

This help was created on 12.03.2004