R-Nb Clust. Can you acces alternative -not best- clustering partitions?
0
0
Entering edit mode
3 months ago

I was doing some clustering trying to see if my i can distinguish groups in my data, and i stumbled uppon NbClust.

But when i run my code, i can obtain the best partition according to majority rule, but is there a way to acces another alternatives partitions? According to majority rule, i can identify two clusters in my data, but there is like 7 indexes that identify 5 clusters as the best option.

Can i access this partitions in any way? I tried using a vector as the index value, but it only takes the first argument. I supose i could go index by index, but i understand that the ideal here would be to look up at the same time to the 7 partitions that those indexes use to make the 5clusters in order to the function to get the best.

res <- NbClust(data[3:ncol(data)],distance="euclidean",min.nc=2,max.nc=8 ,method="ward.D2",index="alllong") 
res <- NbClust(data[3:ncol(data)],distance="euclidean",min.nc=2,max.nc=8 ,method="ward.D2",index=c("rubin","cindex"))

This only takes the first item in the vector as the argument.

R cluster NbClust • 90 views
ADD COMMENT

Login before adding your answer.

Traffic: 2212 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6