在 R USArrests 数据集中切割树状图不会返回正确的树状图

Question

I have been looking online for tutorial but the result is not correct

d <- dist(USArrests, method = "euclidean") # distance matrix
usarrests_hi_cluster <- hclust(d, method="complete") 

plot(usarrests_hi_cluster)

This works fine and I get the whole dendogram

but when I do the following to prune at certain height

my_dend<-cutree(usarrests_hi_cluster, h = 150)

print(my_dend)

only get a list of the states with a number

       Alabama         Alaska        Arizona       Arkansas 
             1              1              1              2 
    California       Colorado    Connecticut       Delaware 
             1              2              3              1 
       Florida        Georgia         Hawaii          Idaho 
             1              2              3              3 
      Illinois        Indiana           Iowa         Kansas 
             1              3              3              3 
      Kentucky      Louisiana          Maine       Maryland 
             3              1              3              1

I want to cut the dendogram at a certain height to get 3 clusters and plot it

I am tryingt to answer this question

(b) Cut the dendrogram at a height that results in three distinct clusters. Which states belong to which clusters?

Answer 1

See this question/answer for the same question but plotting the top of the tree: Plotting hclust only to the cut clusters, not every leaf

What you are looking for are is plotting the lower branches. The branch levels is stored as a list in the "$lower" variable of the data structure.

hc <- hclust(dist(USArrests))
#cut tree
dend2 <- cut(dend1, h = 150)
#cycle through the lower list and plot
lapply(dend2$lower, function(x){plot(x)})

To answer part (b): The output of print(my_dend) is showing which branch each state belongs to.