R语言聚类算法

• 1. 导入并检查数据
• 2. 确定最佳聚类数目
• 3. kmeans聚类
• • 3.1 查看聚类结果
  • 3.2 提取类标签并且与原始数据进行合并
  • 3.3 查看每一类的数目
  • 3.4 进行可视化展示
• 4. 层次聚类
• • 4.1 可视化

#要是没有这个包的话，首先需要安装一下
#install.packages("factoextra")
#载入包
#若载入失败,输入:
#install.packages("caret")
library(factoextra)
# 数据进行标准化
df <- scale(USArrests) 
# 查看数据的前五行
head(df, n = 5)

		               Murder   Assault   UrbanPop         Rape
		Alabama    1.24256408 0.7828393 -0.5209066 -0.003416473
		Alaska     0.50786248 1.1068225 -1.2117642  2.484202941
		Arizona    0.07163341 1.4788032  0.9989801  1.042878388
		Arkansas   0.23234938 0.2308680 -1.0735927 -0.184916602
		California 0.27826823 1.2628144  1.7589234  2.067820292

#确定最佳聚类数目
fviz_nbclust(df, kmeans, method = "wss") + 
+ geom_vline(xintercept = 4, linetype = 2)
#可以发现聚为四类最合适，当然这个没有绝对的，从指标上看，选择坡度变化不明显的点最为最佳聚类数目。

#设置随机数种子，保证实验的可重复进行
set.seed(123)
#利用k-mean是进行聚类
km_result <- kmeans(df, 4, nstart = 24)

print(km_result)

K-means clustering with 4 clusters of sizes 13, 16, 13, 8

Cluster means:
      Murder    Assault   UrbanPop        Rape
1 -0.9615407 -1.1066010 -0.9301069 -0.96676331
2 -0.4894375 -0.3826001  0.5758298 -0.26165379
3  0.6950701  1.0394414  0.7226370  1.27693964
4  1.4118898  0.8743346 -0.8145211  0.01927104

Clustering vector:
略

Within cluster sum of squares by cluster:
[1] 11.952463 16.212213 19.922437  8.316061
 (between_SS / total_SS =  71.2 %)

Available components:

[1] "cluster"      "centers"      "totss"       
[4] "withinss"     "tot.withinss" "betweenss"   
[7] "size"         "iter"         "ifault" 

dd <- cbind(USArrests, cluster = km_result$cluster)
head(dd)

           Murder Assault UrbanPop Rape cluster
Alabama      13.2     236       58 21.2       4
Alaska       10.0     263       48 44.5       3
Arizona       8.1     294       80 31.0       3
Arkansas      8.8     190       50 19.5       4
California    9.0     276       91 40.6       3
Colorado      7.9     204       78 38.7       3

table(dd$cluster)

	 1  2  3  4 
	 8 16 13 13

fviz_cluster(km_result, data = df,
             palette = c("#2E9FDF", "#00AFBB", "#E7B800", "#FC4E07"),
             ellipse.type = "euclid",
             star.plot = TRUE, 
             repel = TRUE,
             ggtheme = theme_minimal()
)

#先求样本之间两两相似性,即相似性矩阵
result <- dist(df, method = "euclidean")
#用hclust聚类
result_hc <- hclust(d = result, method = "ward.D2")
#进行初步展示
fviz_dend(result_hc, cex = 0.6)

fviz_dend(result_hc, k = 4, 
          cex = 0.5, 
          k_colors = c("#2E9FDF", "#00AFBB", "#E7B800", "#FC4E07"),
          color_labels_by_k = TRUE, 
          rect = TRUE          
)