Cluster analysis in R: determine the optimal number of clusters

Question

Being a newbie in R, I\'m not very sure how to choose the best number of clusters to do a k-means analysis. After plotting a subset of below data, how many clusters will be

Answer 1

A simple solution is the library factoextra. You can change the clustering method and the method for calculate the best number of groups. For example if you want to know the best number of clusters for a k- means:

Data: mtcars

library(factoextra)   
fviz_nbclust(mtcars, kmeans, method = "wss") +
      geom_vline(xintercept = 3, linetype = 2)+
      labs(subtitle = "Elbow method")

Finally, we get a graph like: