###########################################################################################

# 2 whiskey 

###########################################################################################

originale<-read.table(file.choose(), header=TRUE,sep="\t")

dataset<-as.matrix(originale[,-c(1:2)])

Type<-originale$type
Sample<-originale$sample


#-- Normalizzazione

dataNorm<-scale(dataset, center = TRUE, scale = TRUE)


#-- Kmeans selezione numero cluster


Tot<-NULL
for(i in c(1:6)) {set.seed(7); km<- kmeans(dataNorm, centers=i,iter.max=50);
Tot<-c(Tot,km$tot.withinss)}


plot(Tot, xlab="K",ylab="Somma dei quadrati delle distanze intra-cluster", pch=16, type="o",
main="Elbow plot")

#--2 o 4  Cluster

KM2<- kmeans(dataNorm, centers=2,iter.max=50)

KM4<- kmeans(dataNorm, centers=4,iter.max=50)


Newdata2<-data.frame(Newdata,KM2=KM2$cluster,KM4=KM4$cluster)

View(Newdata2)

############################


#-- Riduzione del numero di variabili con PCA

PCA<-prcomp(dataset, center= TRUE, scale=TRUE)



#-- Per raggruppamento in base alle prime 3 PC:

datasetPCA<-PCA$x[,c(1:3)]

#--2  Cluster

KM2PCA<- kmeans(datasetPCA, centers=2,iter.max=50)



#-- Plot scores

#-- a colori

Color<-as.factor(Type)
levels(Color)<-c("red","blue")
Color<-as.character(Color)

plot(PCA$x[,1], PCA$x[,2],type="n")
points(PCA$x[,1], PCA$x[,2],col=Color,cex=1.2,pch=16)
text(PCA$x[,1], PCA$x[,2],labels=KM2PCA$cluster,pos=3)
abline(h=0,col="gray")
abline(v=0,col="gray")