clustatis: Perform a cluster analysis of blocks of quantitative variables

Description

Hierarchical clustering of quantitative Blocks followed by a partitioning algorithm (consolidation). Each cluster of blocks is associated with a compromise computed by the STATIS method. Moreover, a noise cluster can be set up.

Usage

clustatis(Data,Blocks,NameBlocks=NULL,Noise_cluster=FALSE,
  Unique_threshold=TRUE,scale=FALSE,
  Itermax=30, Graph_dend=TRUE, Graph_bar=TRUE,
  printlevel=FALSE, gpmax=min(6, length(Blocks)-2),  rhoparam=NULL,
  Testonlyoneclust=FALSE, alpha=0.05, nperm=50)

Value

Each partitionK contains a list for each number of clusters of the partition, K=1 to gpmax with:

group: the clustering partition of datasets after consolidation. If Noise_cluster=TRUE, some blocks could be in the noise cluster ("K+1")
rho: the threshold(s) for the noise cluster (computed or input parameter)
homogeneity: homogeneity index (
rv_with_compromise: RV coefficient of each block with its cluster compromise
weights: weight associated with each block in its cluster
comp_RV: RV coefficient between the compromises associated with the various clusters
compromise: the W compromise of each cluster
coord: the coordinates of objects of each cluster
inertia: percentage of total variance explained by each axis for each cluster
rv_all_cluster: the RV coefficient between each block and each cluster compromise
criterion: the CLUSTATIS criterion error
param: parameters called in the consolidation
type: parameter passed to other functions

There is also at the end of the list:

dend: The CLUSTATIS dendrogram
cutree_k: the partition obtained by cutting the dendrogram for K clusters (before consolidation).
overall_homogeneity_ng: percentage of overall homogeneity by number of clusters before consolidation (and after if there is no noise cluster)
diff_crit_ng: variation of criterion when a merging is done before consolidation (and after if there is no noise cluster)
test_one_cluster: decision and pvalue to know if there is more than one cluster
param: parameters called
type: parameter passed to other functions

Arguments

Data: data frame or matrix. Correspond to all the blocks of variables merged horizontally
Blocks: numerical vector. The number of variables of each block. The sum must be equal to the number of columns of Data
NameBlocks: string vector. Name of each block. Length must be equal to the length of Blocks vector. If NULL, the names are B1,...Bm. Default: NULL
Noise_cluster: logical. Should a noise cluster be computed? Default: FALSE
Unique_threshold: logical. Use same rho for every cluster? Default: TRUE
scale: logical. Should the data variables be scaled? Default: FALSE
Itermax: numerical. Maximum of iteration for the partitioning algorithm. Default: 30
Graph_dend: logical. Should the dendrogram be plotted? Default: TRUE
Graph_bar: logical. Should the barplot of the difference of the criterion and the barplot of the overall homogeneity at each merging step of the hierarchical algorithm be plotted? Default: TRUE
printlevel: logical. Print the number of remaining levels during the hierarchical clustering algorithm? Default: FALSE
gpmax: logical. What is maximum number of clusters to consider? Default: min(6, number of blocks -2)
rhoparam: numerical or vector. What is the threshold for the noise cluster? Between 0 and 1, high value can imply lot of blocks set aside. If NULL, automatic threshold is computed. Can be different for each group (in this case, provide a vector)
Testonlyoneclust: logical. Test if there is more than one cluster? Default: FALSE
alpha: numerical between 0 and 1. What is the threshold to test if there is more than one cluster? Default: 0.05
nperm: numerical. How many permutations are required to test if there is more than one cluster? Default: 50

References

Llobell, F., Cariou, V., Vigneau, E., Labenne, A., & Qannari, E. M. (2018). Analysis and clustering of multiblock datasets by means of the STATIS and CLUSTATIS methods. Application to sensometrics. Food Quality and Preference, in Press.
Llobell, F., Vigneau, E., Qannari, E. M. (2019). Clustering datasets by means of CLUSTATIS with identification of atypical datasets. Application to sensometrics. Food Quality and Preference, 75, 97-104. Llobell, F., & Qannari, E. M. (2020). CLUSTATIS: Cluster analysis of blocks of variables. Electronic Journal of Applied Statistical Analysis, 13(2).

Examples

Run this code


 data(smoo)
 NameBlocks=paste0("S",1:24)
 cl=clustatis(Data=smoo,Blocks=rep(2,24),NameBlocks = NameBlocks)
 #plot(cl, ngroups=3, Graph_dend=FALSE)
 summary(cl)
 #with noise cluster
 cl2=clustatis(Data=smoo,Blocks=rep(2,24),NameBlocks = NameBlocks,
 Noise_cluster=TRUE, Graph_dend=FALSE, Graph_bar=FALSE)
 #with noise cluster and defined rho threshold
 cl3=clustatis(Data=smoo,Blocks=rep(2,24),NameBlocks = NameBlocks,
 Noise_cluster=TRUE, Graph_dend=FALSE, Graph_bar=FALSE, rhoparam=0.5)
 #different Noise cluster thresholds
 cl4=clustatis(Data=smoo,Blocks=rep(2,24),NameBlocks = NameBlocks,
 Noise_cluster=TRUE, Graph_dend=FALSE, Graph_bar=FALSE, Unique_threshold= FALSE,
 rhoparam=c(0.6, 0.5,0.4))