R中树状图的轮廓图

Question

我正在为我的数据使用层次聚类。我想在不同数量的集群中绘制 Silhouette 分数。我搜索了很多帖子，发现我需要使用 pam 进行聚类才能绘制 Silhouette 分数。我在想是否有一种方法可以根据层次聚类结果进行绘图？

这是一个示例数据：

structure(list(safe = c(1.5, 1.5, 2, 1, 1, 1, 2, 1.5, 1, 1, 1.5, 
1, 2, 1, 3.5, 1, 1, 1.5, 1.5, 1, 1, 1, 1, 2, 1.5, 1, 1.5, 1, 
1, 1.5, 2.5, 1, 2, 1, 1.5, 1, 1.5, 1, 1, 1, 2.5, 2, 1, 1, 1, 
2, 1, 1.5, 1.5, 1.5, 1.5, 3, 3, 1, 1.5, 2, 1, 1.5, 1, 1.5, 1, 
1.5, 1, 1, 1, 1.5, 1, 1, 1, 1.5, 1.5, 1.5, 1, 2, 1, 1, 1, 1, 
1, 1, 1.5, 1, 1, 2, 1, 1, 1, 1.5, 1, 1.5, 1.5, 1, 1, 1, 1, 1.5, 
1, 2, 1.5, 1, 1), nhood.soccapi = c(1.8, 1.6, 2.8, 2.2, 2, 3.6, 
3.2, 1.8, 1.6, 1.8, 1.4, 3.8, 1.6, 1, 2, 2.2, 1, 1, 2, 1, 1, 
2.2, 1, 1.4, 1.8, 2, 2.4, 1.8, 1, 2, 2.2, 1.6, 2.2, 1.2, 2, 2, 
1, 2, 2, 2, 2, 1.8, 1.4, 1.6, 1, 2.2, 1.4, 1.6, 2.4, 1.2, 1.4, 
2.4, 2, 2.4, 2, 1.8, 1.6, 1, 1.2, 1.8, 2, 3.4, 2, 2, 1, 2, 1, 
1.4, 2, 2.8, 2, 1, 2, 1.2, 1.4, 2, 1.6, 1.4, 2.2, 2.4, 1.8, 1.4, 
1.4, 2, 2.2, 1, 2, 2.2, 1, 1.8, 2.2, 1.6, 1.2, 1, 2, 1, 1.2, 
2.6, 1.8, 1.8, 1.8), DISORDER = structure(c(1L, 1L, 1L, 1L, 1L, 
1L, NA, 1L, 1L, 1L, 1L, 1L, NA, 1L, NA, NA, 2L, 1L, 2L, 1L, 1L, 
1L, 1L, 2L, 1L, 1L, 1L, NA, NA, 1L, 1L, 1L, 2L, 1L, 1L, NA, 2L, 
NA, 1L, 1L, 1L, NA, 1L, 1L, 1L, 1L, NA, 2L, 1L, 1L, 2L, 1L, NA, 
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 1L, 1L, 1L, NA, 1L, 1L, 1L, 
2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, NA, 1L, 1L, 1L, NA, NA, 1L, 
1L, 2L, 2L, 2L, 1L, 1L, NA, 1L, 1L, 1L, 1L, 1L, NA, 2L, NA, 2L
), .Label = c("0", "1"), class = "factor"), Scho_socialdep1000_3 = structure(c(3L, 
3L, 3L, 3L, 3L, 2L, NA, 3L, 2L, 1L, 2L, 1L, NA, 2L, NA, NA, 1L, 
2L, 1L, 1L, 2L, 2L, 1L, 1L, 1L, 3L, 2L, NA, NA, 1L, 2L, 2L, 2L, 
1L, 1L, 2L, 3L, NA, 1L, 1L, 2L, NA, 2L, 2L, 1L, 2L, NA, 1L, 3L, 
2L, 1L, 1L, NA, 3L, 2L, 3L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 2L, 
2L, 1L, 1L, 2L, 3L, 2L, 1L, 1L, 1L, 1L, 2L, 2L, 1L, NA, 1L, 1L, 
1L, NA, NA, 1L, 3L, 3L, 2L, 3L, 2L, 1L, 2L, 1L, 1L, 3L, 2L, 1L, 
NA, 2L, NA, 3L), .Label = c("0", "1", "2"), class = "factor")), row.names = 100:200, class = "data.frame")

还有我的集群代码：

dist <- daisy(dt, metric = "gower")
cls.env <- hclust(dist, method = "average")

Answer 1

hclust returns 只是表示簇内簇的树状图。剪影分数是在一个特定的聚类上定义的，而不是在所有可能的聚类上定义的。这就是为什么它开箱即用地与 Partitioning Arround Medoids (PAM) 一起使用的原因。然而，在层次聚类中，首先需要通过切割树状图树来决定选择哪个聚类。

这是使用 k=5 簇绘制层次聚类轮廓的方法：

library(tidyverse)
library(cluster)

data <- structure(list(safe = c(
  1.5, 1.5, 2, 1, 1, 1, 2, 1.5, 1, 1, 1.5,
  1, 2, 1, 3.5, 1, 1, 1.5, 1.5, 1, 1, 1, 1, 2, 1.5, 1, 1.5, 1,
  1, 1.5, 2.5, 1, 2, 1, 1.5, 1, 1.5, 1, 1, 1, 2.5, 2, 1, 1, 1,
  2, 1, 1.5, 1.5, 1.5, 1.5, 3, 3, 1, 1.5, 2, 1, 1.5, 1, 1.5, 1,
  1.5, 1, 1, 1, 1.5, 1, 1, 1, 1.5, 1.5, 1.5, 1, 2, 1, 1, 1, 1,
  1, 1, 1.5, 1, 1, 2, 1, 1, 1, 1.5, 1, 1.5, 1.5, 1, 1, 1, 1, 1.5,
  1, 2, 1.5, 1, 1
), nhood.soccapi = c(
  1.8, 1.6, 2.8, 2.2, 2, 3.6,
  3.2, 1.8, 1.6, 1.8, 1.4, 3.8, 1.6, 1, 2, 2.2, 1, 1, 2, 1, 1,
  2.2, 1, 1.4, 1.8, 2, 2.4, 1.8, 1, 2, 2.2, 1.6, 2.2, 1.2, 2, 2,
  1, 2, 2, 2, 2, 1.8, 1.4, 1.6, 1, 2.2, 1.4, 1.6, 2.4, 1.2, 1.4,
  2.4, 2, 2.4, 2, 1.8, 1.6, 1, 1.2, 1.8, 2, 3.4, 2, 2, 1, 2, 1,
  1.4, 2, 2.8, 2, 1, 2, 1.2, 1.4, 2, 1.6, 1.4, 2.2, 2.4, 1.8, 1.4,
  1.4, 2, 2.2, 1, 2, 2.2, 1, 1.8, 2.2, 1.6, 1.2, 1, 2, 1, 1.2,
  2.6, 1.8, 1.8, 1.8
), DISORDER = structure(c(
  1L, 1L, 1L, 1L, 1L,
  1L, NA, 1L, 1L, 1L, 1L, 1L, NA, 1L, NA, NA, 2L, 1L, 2L, 1L, 1L,
  1L, 1L, 2L, 1L, 1L, 1L, NA, NA, 1L, 1L, 1L, 2L, 1L, 1L, NA, 2L,
  NA, 1L, 1L, 1L, NA, 1L, 1L, 1L, 1L, NA, 2L, 1L, 1L, 2L, 1L, NA,
  1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 1L, 1L, 1L, NA, 1L, 1L, 1L,
  2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, NA, 1L, 1L, 1L, NA, NA, 1L,
  1L, 2L, 2L, 2L, 1L, 1L, NA, 1L, 1L, 1L, 1L, 1L, NA, 2L, NA, 2L
), .Label = c("0", "1"), class = "factor"), Scho_socialdep1000_3 = structure(c(
  3L,
  3L, 3L, 3L, 3L, 2L, NA, 3L, 2L, 1L, 2L, 1L, NA, 2L, NA, NA, 1L,
  2L, 1L, 1L, 2L, 2L, 1L, 1L, 1L, 3L, 2L, NA, NA, 1L, 2L, 2L, 2L,
  1L, 1L, 2L, 3L, NA, 1L, 1L, 2L, NA, 2L, 2L, 1L, 2L, NA, 1L, 3L,
  2L, 1L, 1L, NA, 3L, 2L, 3L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 2L,
  2L, 1L, 1L, 2L, 3L, 2L, 1L, 1L, 1L, 1L, 2L, 2L, 1L, NA, 1L, 1L,
  1L, NA, NA, 1L, 3L, 3L, 2L, 3L, 2L, 1L, 2L, 1L, 1L, 3L, 2L, 1L,
  NA, 2L, NA, 3L
), .Label = c("0", "1", "2"), class = "factor")), row.names = 100:200, class = "data.frame")



dist <- daisy(data, metric = "gover")
#> Warning in daisy(data, metric = "gover"): with mixed variables, metric "gower"
#> is used automatically
dendro <-
  dist %>%
  hclust(method = "average") %>%
  # need to put the elements into mutually exclusive sets of clusters
  cutree(k = 5)

silhouette(dendro, dist) %>%
  plot()

请注意k的选择是任意的。选择另一个值会得到不同的结果。

^{由 reprex package (v2.0.0)}

于 2022-03-15 创建

R中树状图的轮廓图

Silhouette plot from dendrogram in R

r

silhouette