Hierarchical Clustering of Dissimilarity Analysis Data Frames

This function combines the dissimilarity scores computed by distantia(), the agglomerative clustering methods provided by stats::hclust(), and the clustering optimization method implemented in utils_cluster_hclust_optimizer() to help group together time series with similar features.

When clusters = NULL, the function utils_cluster_hclust_optimizer() is run underneath to perform a parallelized grid search to find the number of clusters maximizing the overall silhouette width of the clustering solution (see utils_cluster_silhouette()). When method = NULL as well, the optimization also includes all methods available in stats::hclust() in the grid search.

This function supports a parallelization setup via future::plan(), and progress bars provided by the package progressr.

Usage

distantia_cluster_hclust(df = NULL, clusters = NULL, method = "complete")

Arguments

df: (required, data frame) Output of distantia(). Default: NULL
clusters: (required, integer) Number of groups to generate. If NULL (default), utils_cluster_kmeans_optimizer() is used to find the number of clusters that maximizes the mean silhouette width of the clustering solution (see utils_cluster_silhouette()). Default: NULL
method: (optional, character string) Argument of stats::hclust() defining the agglomerative method. One of: "ward.D", "ward.D2", "single", "complete", "average" (= UPGMA), "mcquitty" (= WPGMA), "median" (= WPGMC) or "centroid" (= UPGMC). Unambiguous abbreviations are accepted as well. If NULL (default), utils_cluster_hclust_optimizer() finds the optimal method. Default: "complete".

Value

list:

cluster_object: hclust object for further analyses and custom plotting.
clusters: integer, number of clusters.
silhouette_width: mean silhouette width of the clustering solution.
df: data frame with time series names, their cluster label, and their individual silhouette width scores.
d: psi distance matrix used for clustering.
optimization: only if clusters = NULL, data frame with optimization results from utils_cluster_hclust_optimizer().

Examples


#weekly covid prevalence in California
tsl <- tsl_initialize(
  x = covid_prevalence,
  name_column = "name",
  time_column = "time"
)

#subset 10 elements to accelerate example execution
tsl <- tsl_subset(
  tsl = tsl,
  names = 1:10
)

if(interactive()){
  #plotting first three time series
  tsl_plot(
    tsl = tsl[1:3],
    guide_columns = 3
  )
}

#dissimilarity analysis
distantia_df <- distantia(
  tsl = tsl,
  lock_step = TRUE
)

#hierarchical clustering
#automated number of clusters
#automated method selection
distantia_clust <- distantia_cluster_hclust(
  df = distantia_df,
  clusters = NULL,
  method = NULL
)

#names of the output object
names(distantia_clust)
#> [1] "cluster_object"   "clusters"         "silhouette_width" "df"              
#> [5] "d"                "optimization"    

#cluster object
distantia_clust$cluster_object
#> 
#> Call:
#> stats::hclust(d = d_dist, method = method)
#> 
#> Cluster method   : ward.D 
#> Number of objects: 10 
#> 

#distance matrix used for clustering
distantia_clust$d
#>               Alameda    Butte Contra_Costa El_Dorado   Fresno Humboldt
#> Butte        2.962963                                                  
#> Contra_Costa 1.162055 2.733068                                         
#> El_Dorado    2.483755 2.327273     2.767442                            
#> Fresno       3.456869 3.324759     3.387755  4.433962                  
#> Humboldt     3.960000 4.233871     4.380952  3.568627 4.439863         
#> Imperial     4.119107 4.408978     4.328125  4.568627 3.184685 4.173228
#> Kern         3.658065 2.811688     3.670103  3.942857 2.381766 4.388889
#> Kings        3.631491 3.623529     3.719723  3.694352 2.981191 3.481739
#> Los_Angeles  3.972376 4.005556     3.871720  4.517711 2.302730 4.647059
#>              Imperial     Kern    Kings
#> Butte                                  
#> Contra_Costa                           
#> El_Dorado                              
#> Fresno                                 
#> Humboldt                               
#> Imperial                               
#> Kern         3.696145                  
#> Kings        1.917582 3.203150         
#> Los_Angeles  3.598377 3.420000 2.954876

#number of clusters
distantia_clust$clusters
#> [1] 5

#clustering data frame
#group label in column "cluster"
#negatives in column "silhouette_width" higlight anomalous cluster assignation
distantia_clust$df
#>            name cluster silhouette_width
#> 1       Alameda       1        0.5733007
#> 2         Butte       2        0.1828440
#> 3  Contra_Costa       1        0.5774736
#> 4     El_Dorado       2        0.1136219
#> 5        Fresno       3        0.2402546
#> 6      Humboldt       4        0.0000000
#> 7      Imperial       5        0.4510322
#> 8          Kern       3        0.1410575
#> 9         Kings       5        0.3705427
#> 10  Los_Angeles       3        0.1267346

#mean silhouette width of the clustering solution
distantia_clust$silhouette_width
#> [1] 0.2776862

#plot
if(interactive()){

  dev.off()

  clust <- distantia_clust$cluster_object
  k <- distantia_clust$clusters

  #tree plot
  plot(
    x = clust,
    hang = -1
  )

  #highlight groups
  stats::rect.hclust(
    tree = clust,
    k = k,
    cluster = stats::cutree(
      tree = clust,
      k = k
    )
  )

}

Usage

Arguments

Value

See also

Examples