Its functionality includes exploratory data analysis, data segmentation and data visualization.It is designed to handle realistic data sets : hedonic data set and sensory data set. It makes use of several clustering methods as well as the implementation of partition-validity approach.
Finally, a graphical user interface is implemented with R shiny in order to propose a user friendly package.
You can install the development version from GitHub with :
install.packages("devtools") devtools::install_github("BouzidiImen/ClusteringR")
You can find below an overall look at how ClusteringR can be useful for your sensory analysis.
Diverse methods of clustering are available in the ClusteringR package :
'hierarchical', 'diana', 'kmeans', 'clara', 'pam', 'sota' and 'som'
library(ClusteringR)
library(ClusteringR) # Create a clustering object ------------------------------------------------- cl <- Clustering(t(hedo),ClustMeth='hierarchical',k=3,Hdismethod='euclidean',Hmethod="ward.D2", Graph=F,VarCart=F,IndCart=F ) # get clusters clusters=cl$classes #Plot of dendrogram plot(cl$dendrogram)
help("Clustering") # to see more information about the function of clustering
Based on the sensory map, this package make it easier to know consumers's behaviour, their likes and dislikes.
library(ClusteringR)
library(ClusteringR) # Create an EPM object ------------------------------------------------- E <- EPM(hedo,senso,ModelType='Quadratic',respt=FALSE,nbpoints=50,Graphpred=FALSE,Graph2D=TRUE,Graph3D=FALSE,statistic.Value.Type='rsquared')
help("EPM") # to see more information about the function of external preferences mapping
#Usage library(ClusteringR) S=senso # sensory data H=hedo # hedonic data
Within the package you find a shiny application that demonstrate what the package does and make its use easier.
ClustShiny() #run shiny application
P.S : You can visit the following link to get a sneak peek on the package functionalities.
Shiny application for the package
In preparation of my package, I had to take the help and guidance of my professor Ibtihel Rebhi, who deserves my deepest gratitude.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.