qha (version 0.0.8)

classification: Cluster Analysis

Description

It makes cluster analysis of the data using the nfcl first factorial coordinates

Usage

classification(datos,numclass=2,numaxes=5,pesos=NULL)

Arguments

datos

object of type data frame or matrix with coordinates of factorial analysis

numclass

number of class. Default 2

numaxes

number of dimensions for the classification. Default 5

pesos

a vector of row weights. Default NULL

Value

An object of class "kmeans", a list with components: cluster, centers, withinss and size. To more information see the function kmeans.

Details

The data given by datos is clustered by the Ward method using the function ward.cluster of the package FactoClass. This classification is optimized across the method kmeans.

Examples

data(ratingTV)
 

# Fuzzy Correspondence Analysis 
fuzzyb<-fuzzy(datos=ratingTV$tab, vectorc=c(15,15,15,15,15,15))  
6

#Cluster analysis
clases<-classification(datos=fuzzyb$li,numclass=7, numaxes=6)
clases$cluster
clases$centers
clases$size