Classification Conceptuelle avec Généralisation par Intervalles [Conceptual clustering with generalization by intervals]

• Type: Article in International Scientific Journal
• Date: 2012

Authors

P. Brito (Author) FEP; G. Polaillon (Author) Other

Journal

Title: Revue des Nouvelles Technologies de l'Information

Vol. E No. 23

Pages: 35-40

Indexing

Scopus

Scientific classification

FOS: Natural sciences > Mathematics

CORDIS: Physical sciences > Mathematics > Discrete mathematics

Other information

Abstract (EN): This paper deals with hierarchical or pyramidal conceptual clustering methods, where each formed cluster corresponds to a concept, i.e., a pair (extent, intent). We consider data presenting real or interval-valued numerical values, ordered values and/or probability/frequency distributions on a set of categories. Concepts are obtained by a Galois connection with generalisation by intervals, which allows dealing with different variable types on a common framework. In the case of distribution data, the obtained concepts are more homogeneous and more easily interpretable than those obtained by using the maximum and minimum operators previously proposed. A measure of generality of a concept is defined similarly for all these variable types. An example illustrates the proposed method.

Language: French

Type (Professor's evaluation): Scientific