4. Classification

4.1 Objectives

As in the case of multidimensional positioning (MDS), the data, which concern n individuals assigned, for simplicity, the same weight, can be presented in different forms:

• a table of distances or measures of dissimilarity, n x n, between individuals taken in pairs;

• the observations of p quantitative variables on these n individuals associated with a choice of metric ;

• observations, always on these n individuals, of qualitative variables or a mixture of quantitative and qualitative variables.

In one way or another, in each case, it's a matter of returning to the table of pairwise distances between individuals (i.e. the first case). Choosing...