AbstractThe aim of this paper is to present the concepts underlying an approach to data analysis using a hierarchical classification. The data can be provided by observation, experiment or knowledge. We begin by presenting the classical view of the context of data representation, in which the algorithm of hierarchical ascendant construction of the classification tree is set. The main notion in our method is one of ‘similarity’. The latter must be elaborated in the best way, taking into account the mathematical nature of the objects to be compared. In this elaboration, we adopt a set theoretic and combinatoric representation of the descriptive attributes, which are interpreted in terms of relations. On the other hand, we introduce a probability scale for similarity measurement by using a likelihood conc