# R program for additional multivariate exercise 7 (VHM 802)
# only includes code for K-means clustering

goblet <- read.csv("r:/goblet.csv", header=TRUE, row.names=1)
sgoblet <- scale(goblet)
set.seed(210325)

sgoblet.k2 <- kmeans(sgoblet,centers=2,nstart=1000)
sgoblet.k2

sgoblet.k3 <- kmeans(sgoblet,centers=3,nstart=1000)
sgoblet.k3

sgoblet.k4 <- kmeans(sgoblet,centers=4,nstart=1000)
sgoblet.k4

sgoblet.k5 <- kmeans(sgoblet,centers=5,nstart=1000)
sgoblet.k5

sgoblet.k6 <- kmeans(sgoblet,centers=6,nstart=1000)
sgoblet.k6

ntgoblet <- goblet/(goblet[,1]+goblet[,2]+goblet[,3]+goblet[,4]+goblet[,5]+goblet[,6])
nhgoblet <- goblet/goblet[,3]
sntgoblet <- scale(ntgoblet)
snhgoblet <- scale(nhgoblet)

sntgoblet.k2 <- kmeans(sntgoblet,centers=2,nstart=1000)
sntgoblet.k2
# continue with K=3,4,5...

snhgoblet.k2 <- kmeans(snhgoblet,centers=2,nstart=1000)
snhgoblet.k2
# continue with K=3,4,5...